^^f 'p.ff. ',^ ^. h^: £3S; ■ « i > QH431 D24 fe t (Tlip S. 1. IltU IGtbrarg D24 "A N 'ssnJojAg - _WaNI8 HlHdWvT" ^rr THIS BOOK IS DUE ON THE DATE INDICATED BELOW AND IS SUB- JECT TO AN OVERDUE FINE AS POSTED AT THE CIRCULATION DESK. lOOM/5-79 north C- ■" .\tirary. Rale'.gh Inheritance in Poultry BY C. B. DAVENPORT DIRECTOR OF STATION FOR EXPERIMENTAL EVOLUTIO COLD SPRING HARBOR, NEW YORK ^ ^ *> ^^1 WASHINGTON, D. C Published by the Carnegie Institution of Washington 1906 North Carolina State Library Gift of I North Carolina State Library V/i i - Inheritance in Poultry BY C. B. DAVENPORT DIRECTOR OF STATION FOR EXPERIMENTAL EVOLUTION, COLD SPRING HARBOR, NEW YORK WASHINGTON, D. C Published by the Carnegie Institution of Washington 1906 PUBLICATIONS OF THE CARNEGIE INSTITUTION OF WASHINGTON, No. 52 PAPERS OF THE STATION FOR EXPERIMENTAL EVOLUTION, No. 7 PRESS OF JUDD & DETWEILER, INC. WASHINGTON, D. C TABLE OF CONTENTS. Page A. Statement of Problem i B. Method and Material 5 C. Results of Crossing 6 Series I. Single-comb Black Minorca and White-crested Black Polish 6 Statement of Problem 6 The Races as a Whole 6 Table of Characteristics 6 Remarks on the Characteristics 7 I. Comb, 7 ; 2. Nostrils, 7 ; 3. Form of Skull, S ; 4. Crest, 9 ; 5. Color of Crest, 10. Material 10 Results 10 I. Comb, 10; 2. Nostrils, 12; 3. Cerebral Hernia, 13 ; 4. Crest, 14; 5. Color of Top of Head, 15 ; 6. Correlation of Characters, 16. Conclusions 17 Series II. Single-comb White Leghorn and Houdan 18 Statement of Problem 18 The Races as a Whole i8 Table of Characteristics 19 Discussion of Characteristics 19 I. General Plumage Color, 19 ; 2. Color of Upper Mandible, 19 ; 3. Nostrils, 19 ; 4. Comb, 19 ; 5. Whiskers or Muff, 20 ; 6. Beard, 20 ; 7, 8. Crest ; Cere- bral Hernia, 20; g. Foot Color, 20 ; 10. Number of loes, 20. Previous Investigations 21 Material 21 Results 21 I. General Plumage Color, 21; 2. Color of Upper Mandible, 22 ; 3. Nostrils, 22 ; 4. Comb, 22 ; 5. Face Feathering, 23 ; 6. Beard, 24; 7. Cerebral Hernia, 24; 8. Crest, 24 ; 9. Foot Color, 24 ; 10. Numberof Toes, 25; 11. Corre- lation of Characters, 25. Conclusions 26 Series III. Houdan and Single-comb Black Minorca 27 Statement of Problem 27 The Races as a Whole 27 Table of Characteristics 27 Material 28 Results 28 I. General Plumage Color, 28; 2. Comb, 28 ; 3. Nostril Form, 28; 4. Crest, 28 5. Cerebral Her- nia, 28 ; 6. 7. Muff and Beard, 28 ; 8. Foot Color, 28; 9. Toes, 28. CouclUbions 28 Page Series IV. Single-comb White Leghorn and Rose-comb Black Minorca 29 Statement of Problem 29 The Races as a Whole 29 Table of Characteristics 29 Remarks on the Characteristics 29 I. General Plumage Color, 29 ; 2. Comb Form, 29; 3. Foot Color, 29. Material 29 Results 30 1. Plumage Color, 30; 2. Comb Form, 30 ; 3. Foot Color, 31. Conclusions 31 Series V. Single-comb Black Minorca and Dark Brahma 31 Statement of Problem 31 The Races as a Whole 31 Table of Characteristics 32 Remarks on the Characteristics 32 I. General Plumage Color, 32; 2. Wing Bars, 32 ; 3. Comb, 32 ; 4. Karlobe Color, 33 ; 3. Iris Color, 33 ; 6. Foot Color, 33 ; 7. Foot Feathering, 34 ; 8. Vul- ture Hock, 34. Material 34 Results 34 I. General Plumage Color, 34 ; 2. Wing Coverts, 35 ; 3. Comb, 35 ; 4. Earlobe Color, 35 ; 5. Iris Color, 35 ; 6. Beak and Foot Color, 35 ; 7. Foot Feathering, 35 ; 8. Vulture Hock, 35. Conclusions 36 Series VI. White Leghorn and Dark Brahma 36 Statement of Problem 36 The Races as a Whole 36 Table of Characteristics 36 Remarks on the Characteristics 36 I. Hackle Color, 36; 3. Wing Bow, 37. Material 37 Results 37 I. General Plumage Color, 37; 3. Wing Coloration, 37; 4. Tail Color, 38 ; 5. Comb Form, 38 ; 6. Karlobe, 38; 7. Iris Color, 38; 8. Vulture Hock, 38 ; 9. Foot Feathering, 38. Conclusions .*... 38 Series VII. Black Cochin Bantam and White Leghorn Bantam 39 Statement of Problem 39 The Races as a Whole 39 Table of Characteristics 39 (III) IV TABLE OF CONTENTS. Page Remarks on the Characteristics 39 1. Geueral Plumage Color, 39 ; 2. Earlobe Color, 39 ; 3. Vul- ture Hock, 39. Material 39 Results 40 I. General Plumage Color, 40 ; 2. Earlobe Color, 40 ; 3. Vulture Hock, 40; 4. Foot Feathering, 40. Conclusions 40 Series VIlI. White Le.^horu Bantam and Buff Cochin Bantam 40 Statement of Problem 40 The Races as a Whole 40 Table of Characteristics 41 Remarks ou the Characteristics 41 I. General Plumage Color 41 Material 42 Results 42 I. General Plumage Color, 42 ; 2. Earlobe Color, 43 ; 3. Vulture Hock, 43; 4. Foot Feathering,43. Conclusions 43 Series IX. Tosa Fowl (Yokohama) and White Cochin Bantam 43 Statement of Problem 43 The Races as a Whole 43 Tablt: of Characteristics 44 Remarks on the Characteristics 44 I. General Plumage Color, 44 ; 2. Tail, 44; 3- Foot Feather- ing, 48 ; 4. Foot Color, 48. Material 4S Results 49 I. General Plumage Color, 49 ; 2. Tail Length, 49 ; 3. Foot Feathering, 50 ; 4. Foot Color, 50 ; 5. Correlation of Character- istics, 51. Conclusions 51 Series X. Dark Brahma and Tosa Fowl 51 Statement of Problem 51 The Races as a Whole 51 Table of Characteristics 52 Remarks ou the Characteristics 52 I. Shafting, 52 ; 2. Hackle l^ac- iug, 52; 3- Body Lacing, 52; 4. Penciling, 53 ; 5. Red Wing- Bar, 53 ; 6. White Wing-Bows, 53 ; 8. White Earlobe, 53 ; 9. Iris Color, 53. Material 53 Results 54 I. Shafting, 54; 2. Hackle Lac- «"!?. 54 ; 3- Body Lacing, 54 ; 4. Peuciliug, 54 ; 5. Red Wing- Bar, 54 ; 6. White Wing-Bow, 54; 7- Comb, 54; 8. Earlobe Color, 54 ; 9. Iris Color, 54 ; 10. Foot Color, 54; II. Vulture Hock, 54; 12. Foot Feathering, 55 ; 13- Tail Feathers, 55- Page Conclusions 55 Method of Inheritauce, 55 ; Sex in Inheritance, 55. Series XI. Frizzle and Silky 55 Statement of Problem 55 The Races as a Whole 55 Table of Characteristics 57 Remarks ou the Characteristics 57 I. Plumage Color, 57 ; 2. Comb Form, 57; 3-5. Feather Form, 57 ; 6. Number of Toes, 59 ; 7. Skin Color, 59. Material 59 Results 59 I. Plumage Color, 59 ; 2. Comb, 59; 3-5- Curvingof Shaft, Barb Length, and Barb Form, 59 ; 6. Numl)er of Toes, 60 ; 7. Skin Color, 60 ; 8. Crest, 60. Conclusions 60 Series XII. Single-comb White Leghorn Bantam and Black-breasted Red Rump- less Game 61 Statement of Problem 61 The Races as a Whole 61 Table of Characteristics 61 Remarks on the Characteristics 61 4. Uropygium 61 Material 62 Results 62 I. General Plumage Color, 62 ; 2. Beak Color, 63; 3. Uropygium, 63 ; 4. Foot Color, 63. Conclusions 63 Series XIII. Black Cochin Bantam and Black-breasted Red Rumples.s Game 63 Statement of Problem 63 The Races as a Whole 63 Table of Characteristics 63 Material 63 Results 64 I. General Plumage Color, 64; 2. Uropygium, 64; 3. Iris Color, 64 ; 4. Vulture Hock, 64 ; 5. Foot Feathering, 64. Conclusions 64 D. Geueral Discussion 65 Inheritance of Particular Characteristics... 65 Comb Form 65 Nostril Form 68 Cerebral Hernia 69 Crest 69 Whiskers, or Muflf. 70 Beard 70 Feather Form 70 Uropygium 71 Tail-Length 71 Vulture Hock 71 Foot Feathering 72 Extra Toes 72 Skin Color 73 Mandible Color 74 Foot Color 74 TABLE OF CONTENTS. Page Iris Color 74 Earlobe Color 74 General Plumage Color 74 White vs. Dark 75 Dominance of White 75 Barring 75 Andalusian Coloration 76 White vs. Buff 76 Black z/j. Red 76 Color of Top of Head 76 Color of Hackles — Hackle I,acing 77 Wing Color— Red Wing Coverts 77 Tail Color 77 Shafting 77 Body Lacing 78 Penciling 78 General Topics in Inheritance 78 Unit Characters 78 Alternative, Particulate (Mosaic), and Blending Inheritance 81 Inheritance of Specific vs. Varietal Characteristics 82 Inheritance of Positive vs. Negative Varietal Characteristics 83 Inheritance of Old vs. New Character- istics 84 Dominance and Recessiveness 84 • Page Dependence of Dominance on the Races Crossed 86 Prepotency and Dominance 87 Hybrid Forms 88 Reversion 90 Purity of Gametes 91 Comparison of Reciprocal Crosses 93 Inheritance of Sexually Dimorphic Characteristics and Sexual Dimor- phism in the Hybrids 93 Black Minorca and Dark Brahma... 94 White Leghorn and Dark Hrahma.. 94 White Leghorn and Houdan 94 White Leghorn and Rose-comb Black Minorca 95 Tosa Fowl and White Cochin Ban- tam 95 Dark Brahma (female) and Tosa Fowl (male) 95 Transfer of Sexually Dimorphic Char- acteristics from One Sex to the Other. 95 Sex in Hybrids 97 Correlation of Characteristics 97 The Mutation Theory in its Relation to the Origin of Domesticated Anima's.. 98 E. Summary of Conclusions 100 F. Literature Cited loi INHERITANCE IN POULTRY. By C. B. Davenport. Evolution proceeds by steps. These steps are measured by the character- istics of organisms. When in the evolution of a race a characteristic is added a progressive step is taken. When a characteristic drops out a retro- gressive step is made. Since the characteristic is the unit of evolution, it deserves careful study. The present work is a first study of the method of inheritance of characteristics. A. STATEMENT OF PROBLEM. When by some abnormal process a single, fertilized egg develops into two individuals they are, and continue throughout life to be, almost indistin- guishable. This holds true even when the conditions of life of the two are dissimilar. This case is exemplified by ' ' identical twins ' ' as they occur in man,* The great similarity of such identical twins teaches that environ- ment plays a small part in determining adult characteristics as compared with heredity. Consequently more confidence can be felt that the results of hybridization experiments are directly due to inheritance ; they are little affected by varying environment. The children of the ordinary family are not identical in appearance, although showing marked family traits. Certain characteristics may be common, but others are peculiar to each individual child. This proves that the fertilized eggs of the same two parents have not the same hereditary potentialities. It indicates also that we cannot predict the characteristics of the offspring from those of the parents. The proportion of qualities derived from either one of the two parents will differ in different children, or new qualities may appear. This is because the offspring do not inherit from the visible part of the parents' bodies but from their hidden germ cells or ' ' gametes. ' ' And the characteristics of the soma are never through- out the same as those of the ripe gametes it carries. When the parents belong to different races having markedly dissimilar characteristics there is not merely the question of dissimilarity of the off- spring but of the inheritance of the antagonistic characteristics. Until * Gallon, F., 1883, pp. 216-243. Compare also for a critical study of resemblance in twins, Thorndike, 1905. 2 INHERITANCE IN POULTRY. recently the law has been commonly accepted which is thus expressed by Darwin (1876, Chapter XV) : "When two breeds are crossed their char- acters usually become intimately fused together." Many cases of non- fusing inheritance are now known and it is important to ascertain the rela- tive frequency of the different kinds of inheritance and their relation to one another. Lucas (1850, p. 194) recognizes three methods of inheritance, which he calls respectively that of election, of mixture, and of combination. They are thus defined : Election results in imprinting on some or all parts of the organism the characteristics of the father exclusively or those of the mother. Mixture results in a mixed or simultaneous representation of the father and of the mother on some or all of the parts of the organism. Its extreme is fusion of characteristics. Combination results in the substitution of a new characteristic in the place of any representative in a part or over the whole of the organism. This new characteristic results from the interaction of the two antagonistic ones just as a chemical combination often differs wholly from the elements which have been united in its manufacture. Darwin (1876, Chapter XV) seems to recognize only two classes of inherit- ance, yiz. , one in which characteristics blend and one in which they refuse to blend. Of the latter class, however, there are two cases ; either the hybrid receives all its characters from one of its parents only, or the hybrid receives part of its characters from one parent, the rest from the other. Nageli (1884; 1898, p. 17) describes the different forms of inheritance very clearly, thus : In the idioplasm of a germ cell arising from the crossing of unlike individuals the micellar rows of the individual Aulagen have sometimes an intermediate constitution and produce characteristics in the organism which are intermediate between the char- acteristics of the parents. Sometimes the micellar rows derived from the father and the mother respectively lie side by side interchanged in the idioplasm of the offspring in distinct groupings and may reproduce in the organism their respective characteristics side by side, or onl)' one of them may develop, while the other remains latent. (Clark's translation.) Galton (1889, pp. 7, 12, 14) distinguishes three kinds of inheritance, as follows: (i) Partiailate, or inheritance "bit by bit, this element from one progenitor, that from another ; " (2) blending, as in human skin color ; this may "be none the less 'particulate' in its origin, but the result may be regarded as a fine mosaic too minute for its elements to be distinguished in a general view;" and (3) exclusive, as in human eye color; although "there are probably no heritages that perfectly blend or that absolutely exclude one another, but all heritages have a tendency in one or the other direction, and the tendency is often a very strong one." The different types of inheritance are thought by various authors to be characteristic of particular sorts of crossing. Isidore Geoff roy St.-Hilaire insisted ' ' that the transmission of characters without fusion occurs very STATEMENT OF PROBLEM. 3 rarely when species are crossed." De Vries (1905, pp. 253, 280) concludes that blending and particulate inheritance of qualities characterize the offspring of crossed species, whereas an alternative inheritance of qualities is characteristic of the offspring of a species crossed with a variety * or of two varieties crossed iiiter se. In the case of alternative inheritance there often is exhibited an extremely suggestive phenomenon. When hybrids showing such inheritance are crossed i7itcr sc there is a segregation of the various alternative character- istics into different individual offspring. This is the discovery of Mendel (i866).t The attempt has naturally been made to generalize Mendel's law — to make it apply universally. In my own study the applicability of this law has been kept constantly in mind. * It is to be recalled that in the De Vries system a variety differs from its parent species either in that a characteristic of the species has become latent in the variety or in that a characteristic vehicli was latent in the species has reappeared in the variety. A new species, on the other hand, differs from its parent species in the acquisition of one or more wholly new characteristics. "In normal fertilization and in the intercrossing of varieties all characters are paired." Hence the paired characters struggle together in the zygote and the stronger one of the pair dominates or covers over the weaker one. Thus inheritance is alternative or exclusive. "In crosses between elementary species the differentiating marks are not mated." Hence there is no such struggle between characteristics ; consequently those of both parents reappear in the offspring, iuterdigi- tating. fThe rediscovery of Mendel's work simultaneously by De Vries and by Correns in 1900 will always rank as one of the interesting coincidences in the history of science. There is evidence that others had independently discovered this law in their own work in the last third of the nineteenth century, but the history of this law is still to be written. I may note that Haacke, in 1S93, as a result of extensive breeding of animals, expresses the law of purity of the germ cells. He has the theoretical idea that inheritance is con- veyed both by the plasma (P) and the nucleus (Kern, K). In the union of dissimilar races two kinds of plasma (P and P^) and two kinds of nuclear material (K and K^) may be distinguished. On page 236 he says : Die beideu verschiedenen Plasmen P und P' die sich bei der Befruchtuns vereiuigt haben, trenuen sich wieder bei der Reductionsteiluug der Keimzelle, uud dasselbe gilt von den beiden Kernstoffe K und K'. Diese Trennuiigist in manchen Fallen, wie es scheint, eine vollige, so dass die Plasmen und die Kernstoffe, ab^esehen von den mehr oder minder weitgehenden, aber niemals vollkommenen Ausgleichunsen ihrer Eigenschafien, diedurch gegeuseitige Beeiufliissung stattfinden miissen, ebeuso rein aus der Vereiuigung hervorgehen, als sie in diese hineingetreten sind. Still further Haacke recognizes that in the separation of qualities that occur in the reduction period of the hybrid germ cells, those from different parents may gather into one germ cell. Since this occurs in accordance with the laws of chance (worked out in an example b}' Haacke), we have various combinations of characteristics in the second hybrid generation. Because of the purity of the germ cells it will often happen that mice having certain qualities will, when bred together, produce only those qualities, however complex their ancestry. For example, white dancing-mice bred together will produce nothing but white dancing-mice. Haacke's results seem to have been overlooked by recent experimenters. 4 INHERITANCE IN POULTRY. In typical Meudelian cases not only do the qualities segregate in the second hybrid generation, but in addition, in the first generation, when two con- trasted characteristics are bred together one of the two is patent in the offspring ; the other does not appear. The first is the dominant quality ; the second is recessive. It seems at first to have been assumed that when one of two antagonistic characteristics was dominant over the other it was so in all cases. Recent studies have, however, greatly expanded our notion of dominance and reces- siveness. Even in alternative inheritance we have to admit various addi- tional phenomena of which the following are examples : Prepotency of a character, elsewhere recessive, in some individual or strain. Latency, as Castle (1905, p. 24) uses the term, or the inactive persistence of a normally dominant characteristic in a recessive individual or gamete. When the recessive is cross-bred the latent characteristic may appear as a dominant. Reversion, or the assumption of an atavistic character by a heterozygote. This is illustrated by the case of the cross between albino and black-and- white mice which throw gray. However, this instance may be one of latency. In this study attention will be paid to these phenomena. What determines dominance in any case ? This is a disputed point. De Vries (1905, pp. 278, 280) suggests " that hybrids between a species and its retrograde variety will bear the aspect of the species," and "that the older character dominates the younger one." However, he says it is not the systematic relation of the two parents of a cross that is decisive, but only the occurrence of the same quality, in the one in an active, and in the other in an inactive condition. Hence, whenever this relation occurs between the parents of a cross the active quality prevails in the hybrid, even when the parents differ from each other in other respects so as to be distinguished as systematic species. Correns (1905) also cites cases in which the active allelomorph dominates. In my studies constant attention is directed toward this matter. To recapitulate : This study has been undertaken to determine the differ- ent forms of inheritance (alternative, particulate, blending) occurring in poultry, and to study especially the phenomena of alternative inheritance as exhibited in this group in order to see in how far they accord with Mendel's law and in how far the accessory phenomena of dominance, latency, and reversion occur. METHOD AND MATERIAL. 5 B. METHOD AND MATERIAL. To answer in the shortest time the foregoing questions about inheritance it was necessary to use some rapid and fecund breeder and to interbreed both varieties and species. But species-breeding is slower and more diffi- cult and not more important than breeding races ; for while on the one hand it may be urged that races are artificial, having arisen under domestication, on the other hand hybridization between established species probably plaj'S little part in nature. What must occur again and again in nature is the mating of a mutation or newly arisen race with the parent species. It has been urged that, in such cases, the rare mutation must be swamped by intercrossing with the numerous representatives of the species. But if new characters do not blend in breeding we can see that a new characteristic once arisen may not be swamped. Consequently the study of inheritance in races assumes first importance, and domestic races afford the best material for such study. Again, if we accept the doctrine that man is a single species, all the mo- mentous questions of human inheritance are questions of race inheritance. The outcome of such an admixture of races as is going on in America is a question of race inheritance. The offspring of a man and a woman having one or more diverse characteristics will follow the laws deduced from a study of crossed races. These are practical problems of human evolution, and experiments made with domesticated races can throw light upon them. The main material utilized has been, as stated, poultry. Poultry offer these great advantages : That they are easily bred in great numbers, that two generations can be reared to maturity in a year, that the}'- stand much inbreeding without loss of fertility, and that the number of well-defined characteristics in the group is very great. In my experiments I have kept 29 pens, each with its cock and one or more hens. To separate the eggs of the different females, trap nests were used in the later experiments to hold the hen until she is released. Her number is read and written, with date, upon the egg. Before placing the eggs in the incubator, a list is made of them. Before hatching, eggs of each of the different parentages are separated into a compartment by them- selves, so that the exact parents of each chick may be known, A legband is applied to the chick the moment it is removed from the pedigree tray of the incubator. By these means I have gained in one year 1,500 offspring derived from known parents, and have reared about 500 of them to a period when their adult plumage characters were distinguishable. For keeping records I have used a field pocketbook and a day book at my work-table. "Loose leaf" forms were used for the description of each of the stock, for its egg record, for a chart of its plumage, and for its photographs. Finally, the results of each set of experiments are kept in a large book, posted nearly to date. INHERITANCE IN POULTRY. In treating of my experiments I propose first to give the results by the races crossed, and, secondly, to discuss in order the problems that were set at the beginning. My experiments led me to lay little stress on the races as named by fanciers. In fact it is not races that have been crossed but characteristics. However, as the breeders' names have a utility in bringing to the mind a certain combination of characteristics, they have been freely employed. The different races whose offspring are discu-ssed in this paper are given below in the order adopted in Wright's Poultry Book. Buff Cochin (Bantam). Biaik Cochin (Bantam). Dark Brahma (Bantam). Biack-breasted Red Game. Single-comb Black Minorca. Rose-comb Black Minorca. vSingle-comb White Leghorn. White-crested Black Polish. Houdan. Frizzle. Silky. Tosa fowl, or Yokohama. Rumpless Game Bantam. C. RESULTS OF CROSSING. Series -Single-comb Black inorca and White-crested Black Polish. STATEMENT OF PROBLEM. The cross was undertaken primarily to learn the method of inheritance of the crest, cerebral hernia, and comb of the remarkable Polish fowl. THE RACES AS A M'HOLH. The Black-crested White Polish (figs, i and 2) belongs to a class — Polish fowl — which is one of the fundamental types of poultry. The origin of the Polish fowl is obscure. They were mentioned by Aldrovandi in the six- teenth centur)^ They are found to-day in most parts of the world, and their most characteristic feature may, indeed, have originated independently many times. This feature is a cerebral hernia and its associated crest of large feathers. The Single-comb Black Minorca (figs. 3 and 4) is a typical representative of the Mediterranean class of poultry — tall, stately, close-feathered, non- broody fowl, — modern representatives of the ancient Egyptian poultry. They seem, indeed, to have come from Spain, those imported to England having, according to Wright (1902, p. 391), come from the island whose name they bear. The single combed form is the original and typical variety. TABLE OF CHARACTERISTICS. No. Characteristic. Single-comb Black Minorca. White-crested Black Polish. I 2 3 4 Comb Nostrils Top of head— skull. Top of head — plumage form. . Top of head— plumage color. Single, very large Narrow Normal Plain Two papillae. Wide or high. Cerebral hernia. Crested. 5 Black While and black. SINGI^ 50 J. [Median X "0-'«^- J ^'(^^/zVf/^a- x no-^///^^^^^^ "^^") i No-splitting X no-splitting. } 37/2%. . ,^T ,. ^, ( Splittim: y. splitting- ^% \ ro.^/ & 25 % {^o-medtan X "o-/«^- I ^,^^^,7^ •„,. ^ ^o-splitting) .. i2>^ j '^'^>- "^^" ) i 'No-splitting X no-splitting. .. ass. Relative frequency (a) 3 {b) I \c, 6 {d) 2 {e) 3 (/) I Summing up, out of every 16 individuals we should expect: Characteristic and class. Relative frequency. Per cent. Median and solit comb ia] -\- ic] 9 3 3 I 56.25 18.75 } 25.00 Median and unsplit comb {b)-\-{d) . . . . No-median and split comb [e) No-median and no-split comb [/) * For further discussion of the y-shaped comb see page 65. 12 INHBRITANCE IN POULTRY. On account of the difficulty of deciding iu the case of any young chicks whether ' no-median and no-sphc comb ' is present, the last two classes are combined in the right-hand column of this table and in columns a and/3 of the table above. In comparing the fit of the expected percentages on the two hypotheses with the actual, it is seen that hypothesis a is the better. However, the real test will come in the F3 generation. On hypothesis a the single-combed individuals bred together should produce only median-combed offspring. On hypothesis ? they should produce some without median comb.* First generation hybrids crossed with Minorca. When the Y-shaped comb is crossed back on the single comb of the Minorca (No. 26 c? ) the following distribution of comb characters was obtained : Characteristic. Single comb Cleft comb Papillae or "absent Frequency. Per cent. 21 ti9 o 52.5 47-5 0.0 This result accords well with the expectation that 50 per cent of the combs shall be of the pure Minorca type and 50 per cent of the heterozy- gous type. 2. Nostrils. — First generation. No case occurred of a typical high nostril; this characteristic is dominated by the narrow nostril ; but this dominance is imperfect. In three cases (5 per cent) the nostril is recorded as one-half high, /. e., having an aperture one-half as high as the extreme. In the other cases, placed in class 2, the breadth w^as less, but still evidently influ- enced by the germinal representative of the ' ' high ' ' characteristic. In two skulls that were dissected the processus nasalis of the premaxillary bone was present. class. Characteristic. Narrow a. One-third high 7 b. One-half high 3 High Total . First generation. Class. /• Per cent. 47 } ^° 83 17 I 3 - 100 Characteristic. Narrow a. One-third high 16 b. One-half high 6 High or nearly high.... Total Second generation. f. Per cent. 45 22 18 85 529 25-9 21.2 100. Second generation. In the second generation the high nostril appears again in full or nearly full size in 21 per cent of the cases. Class 3 includes two *This topic is discussed again, generally, at page 68. t Including one thick comb with a median ridge in addition to the cleft comb, forming 103. a typical pea comb. Seen in unhatched chick XVIII, SINGLE-COMB MINORCA AND WHITE-CRESTED POLISH. 13 cases in which the records read three- fourths high ; but even in the PoUsh fowl the high nostril varies somewhat. Theoretically, we should expect 25 per cent of the second generation to have the recessive high nostril. The deficiency may indicate merely insufficient numbers, or perhaps some of the " one-half high " group truly belong in " class 3." The essential facts are, first, that high and narrow are segregated, and that in the second generation also dominance is frequently imperfect. First generatioyi hybrid crossed "with Minorca. The heterozygous mothers all had narrow nostrils, as of course had the Minorca father. No true case of high nostril occurred. The recorded distribution is as follows : Class. Nostril. No of individuals. I 2 Narrow. ... 12 13 One-third to one-half hisrh It seems probable that the 1 2 individuals with narrow nostrils belong to homogametous birds and the 13 individuals of class 2 to heterogametous birds. 3. Cerebral Hernia. — First hybrid generation. Every bird was without a typical cerebral hernia. Nevertheless some of them showed clear traces of their mixed ancestry. On the frontal bone of all fowl is the so-called frontal eminence which is covered by fascia and the thickened skin of the crest. The profile of the skull from the apex of this eminence to the parietal is nearly straight, except for a slight concavity limiting the frontal eminence. In hybrids this concavity is frequently still more marked, the eminence being higher. Indeed, in one case (No. 405 9 ) the concavity is so marked that, as far as one can judge from the living bird, there is a slight hernia. We see, then, that though plain head is dominant it is incom- pletely so. Second hybrid generation. The following is the distribution of this char- acteristic when the hybrids are bred i7iter se. Expectation is: 75 per cent without hernia, 25 per cent with hernia. The result agrees closely with expectation. Characteristic. Hernia absent. Hernia present Total .. Frequency. 75 23 98 Per cent. 76.5 23-5 lOO.O First hybrid generation crossed with Minorca. Since the first generation is DR and Minorca is D, half of the offspring will be pure dominants and half 14 INHERITANCK liN POULTRY. heterozygous, both halves showing the domiuaut "absence of hernia." The result agrees with expectation. Characteristic. Frequency. Per cent. Hernia absent ^4 TOO Hernia oresent 'i ::: Total 34 1 4. Crest. — First hybrid generation. The crest is present in every hybrid old enough to show a crest, yet always in reduced size. Crest is dominant, but the dominance is imperfect. The crest is larger in the females (fig. 5) than in the males (fig. 6). Characteristic. Frequency. Per cent. Crest absent Crest present Total 70 100 70 Secojid hybrid generation. All records, from embryo chicks as well as adults, give : Characteristic. Frequency. Per cent. Crest absent 23 52 *3o.7 693 Crest present Total 75 loo.o *Fig. II. Expectation is that crest will be absent in 25 per cent of the cases. The excess is probably due to the fact that, since crest develops late, it was noted as absent when it would have appeared in adtilt life. To test this I give a table based on hatched chicks only : Characteristic. Frequency. Per cent. Crest absent II 41 21.2 78.8 Crest present Total 52 lOO.O This gives a close approximation to expectation, with a slight excess of crested individuals. The exact statistical proportion, with its possible errors SINGIvE-COMB MINORCA. AND WHITE-CRESTED POLISH. 15 of cla.ssification, is less essential than the fact of reappearance in about one- fourth of the cases of the recessive characteristic. First hybrid generation crossed with Minorca. Since plain-head is R, the cross is of the order DR X R ; from which we should expect an equal number of heterozygous (crested) and pure recessive (plain-headed) offspring. The result, though based on small numbers, accords with expectation. Characteristic. Crest absent (RR). Crest present (DR) Total.. All records. 12 8 Hatched chicks only. 6 6 20 12 5. Color op Top of Head. — First hybrid generation. All records give : Characteristic. Wholly black . . Black and white Total .... Frequency. 36 20 Per cent. 64-3 35-7 56 loo.o It is to be noted, first, that the white of the crest tends to disappear in the later molts, some birds which showed it at 2 months losing it by 6 months, or showing white at the tip only of the crest feathers. Further, with two exceptions, all crests with white feathers belong to females (which have larger crests than males). The two exceptional males are still young and have only a trace of white remaining ; this will probably disappear in the next molt. Third, the proportion of white to black in the crest is always small — much smaller than iu the Polish crest. Tne result looks like an imperfect dominance of black. Second hybrid generation. Hatched chicks only give : Characteristic. Wholly black . . . Black and white. Total . . . Frequency. Per cent. 24 27 47.1 529 51 loo.o I interpret this irregular result to be due to the imperfect dominance of black. Twenty-five per cent of the individuals have wholly black gametes and 25 per cent wholly black-and-white, or mosaic, gametes. The 50 per cent with mixed gametes tend to be black, but contain white in varying pro- portions. Something more than one-fourth of the black-and-white headed individuals are males. 1 6 INHERITAXCK IN POULTRY. First hybrid crossed with Minorca. All records give : Characteristic. Frequency. Per cent. Wholly black 21 2 91-3 8.7 Black and white Total 23 lOO.O Expectation, assuming complete dominance of black, is lOO per cent of black individuals. Result shows incomplete dominance. Of the black-and- white headed individuals, one is a female ; the other died too early for the sex to be determined. Here, again, dominance is less perfect in the female. 6. Correlation of Characters. — In the Single-comb Black Minorca and in the White- crested Black Polish there is an assemblage of characters that are nearly always associated in those races. The first hybrids have another constant association unlike either of the parents, viz., split comb, black crest (in the male) without cerebral hernia, and low to medium nos- trils (fig. 6). In the second generation of hybrids, on the other hand, occur combinations of characters both of the parental species and also of the first generation of hybrids. These combinations are of the most varied sort, so that characteristics always found associated in one parent species may here be found dissociated. When hybrids are bred i>iter se the following com- binations are obtained : No. 1 Comb. I ^ c 2 j 3 i ' 4 r Y 1 5 1 1 6 i 7 8 J 9 1 lO 1 II 12 13 1 - 14 1 15 ; 1 i6 J 1 17 i " i8 1 19 20 - O . 21 1 22 23 i 24 J Crest. Present . Absent Present . Absent . Pre.'eut. Absent . Present Absent . Present . Absent . Present . Absent . Hernia. Nostril. Absent . Absent . Present Present. Absent . Absent . Present . Present. Absent . Absent . Present. ; Present. ■{ Narrow. High... Narrow . \ High... f : Narrow. I High... f ! Narrow, t High... f Narrow. \ ! High . . . / Narrow . High . Narrow . High... Narrow . High... Narrow . High . . Narrow . High Narrow High... Narrow . High... Number of cases. Actual. 17 o 8 o I o 3 o 8 o 12 O 2 O O o I 8 o I 4 5 o o Calcu- lated. 15 5 5 1.6 5 1.6 1.6 0.5 7-3 2.5 2.5 0.8 2.5 0.8 0.8 0-3 7.3 2.8 2.5 0.8 2.5 0.8 0.8 0.0 Ijorth Carolina State Library Raleigh SINGI.E-COMB MINORCA AND WHITR-CRESTED POLISH. 17 This table gives the distribution of characteristics in 70 individuals. Grouping the individuals under certain alternative characters, we have the following relations of actual and calculated frequency of occurrence of each characteristic : Characteristic. Actual. Expected. Characteristic. Actual. Expected. Split comb. , 29 22 19 49 21 35-0 1 17-5 17-5 52.5 17-5 ' Hernia present .... Hernia absent Nostril hij^h Nostril low 16 54 14 56 17-5 52.5 17-5 52.5 Single comb Papillae Crest present Crest absent The actual never deviates far from the expected. If, however, we compare the actual number of cases of each of the combi- nations with the calculated the result is instructive. For example, in the absence of correlation of characters we should expect a high nostril to be associated with a single comb in 5 or 6 of the 22 cases ; but it is never found so associated. In fact a high nostril never occurs in this cross dissociated jrotn a rudimentary comb. On the other hand, it appears that a low nostril may be associated with a rudimentary comb, but in unexpectedly few cases, 4 instead of about 14. Two of the 4 records are from embryos in the shell, in which therefore adult characteristics were not fully developed and the other two cases are recorded as one-half high. It is quite possible that an atypical nostril and absence of true comb are always associated (fig. 11). In order not to complicate the table too much, the correlation between crest and color of the crest feathers was omitted. A subsidiary table is con- sequently given here : /. „ ^ . f Black 18 Crestpresent j White and black 23 Crest absent { Whit^e and "black :::: l! ■.::■.■.;■. i'. . 6 56 Whether crest is present or absent white occurs on the head ; but it is more apt than not to occur when the crest is present and less apt than not when the crest is absent. CONCLUSIONS. In the cross under consideration no characteristic is inherited in strictly Mendelian fashion, for in no case is dominance complete. The nearest ap- proach to typical Mendelian dominance is exhibited, in the present cross, only when crest is crossed with no crest. The new additive characteristic — crest — 2 l8 INHERITANCE IN POULTRY. is dominant. But the crest of the first generation hj'brids is always of small size. Likewise, plain head is dominant over cerebral hernia, but some of the hybrids have exceptionall}' high frontal prominences. The white color of crest is recessive in the male hybrids, but is not entirel}' shut out from the females. The high nostril is recessive, yet the presence of its representative in the hybrid gives the latter abnormally wide nostrils. Finally, the comb affords us a case of an organ in which neither parental form can be said to be dominant without such an extension of the term as to render it quite vague. Ever}'' individual shows a modified comb — the Y or shaped comb. This is a new form — a heterozygous form — that probably reappears in the heterozygotes of each generation. The facts of correlation show that crest is by no means dependent on cerebral hernia. At the same time I doubt if the absence of present corre- lation disproves the hypothesis that the crest was the result of the hernia. It is at least conceivable that a characteristic that arose as a response to the stimulus of an abnormal ontogenesis should become hereditary and inde- pendent of the stimulus. As for white color on the top of the head, it is dissociable from the crest, for wholly black-crested second hybrids occur. Series II.— Single-comb White Leghorn and HouJan. STATEMENT 01' PROBLEM. This cross was undertaken for comparison with that between Minorca and Polish, and to test the inheritance of plumage color, extra toe, and face feathering. THE RACES AS A WHOLE. The Leghorn (fig. 15) is typical of the Mediterranean class of poultrj^ — slender, tall-legged, close- feathered, nervous, and non-broody — the same class as that to which the Minorca belongs. The white Leghorns came originally from northern Italy.* They have been bred in America since 1834. The single-comb variety is one of the most widely bred of our races and has the reputation of being the greatest egg-producer. Considering its white plumage, its transparent skin, with a trace of yellow, and its red iris, it comes very near to being an albino race, but the retina is pigmented. The Houdan (fig. 16) comes from France. It, like the Dorking, has doubtless descended from the 5-toed fowls of the Romans, described by Columella, which they probably carried to Gaul in their conquest of that * Wright, 1902, p. 385; WyckoflF, 1904, p. 788. ^lyi SVnHB)RAWN SINGLK-COMB WHITE LEGHORN AND HOUDAN. 19 couutr3^ This may have been crossed with ' ' the old crested race of Caux."* The Houdan may be regarded as one of the fundamental types. TABI.E OF CHARACTERISTICS. No. Characteristic. Single-comb White Leghorn . Houdan. I 2 3 4 5 6 General plumage color .... Color of upper mandible .... Nostrils . Comb Face feathering Chin feathering White Yellow Narrow Single (rarelv cleft behind. No. II.) Plain Plain Plain Plain . . Yellow Four Black, white-tipped. Light horn. High. 2-pronged or V. Whiskered. Bearded. 7 8 9 10 Dorsal head plumage Dorsum of cranium Foot color Number of toes . . . Crested. Domed. White. Five. DISCUSSION OF CHARACTERISTICS. 1. General Plumage Color. — In the Leghorn this is typically white, and the most highly selected birds are without trace of black specks or yellowish lacing. The yellow lacing is hard to get rid of. The Houdan color consists typically of black feathers occa.sioually tipped with white (fig. 16). 2. Color op Upper Mandible. — The clear yellow of the mandible of the white lyCghorn is part of the general pigmentation of the skin. Much yellow pig- ment is deposited over the body. It shows prominently in the tarsal scutes. The Houdan mandible is clear black. comb nostril of Fie. B. — Dorsal view of beak of Houdan gAcf showing pail of clublike papillae, c, that represent the v-comb, c. /., cul- minal fold. 3. Nostrils. — The high nostrils of the Houdan (fig. 12) are like those of the Polish fowl (page 7). 4. Comb. — The comb of the Houdan in America is the so-called V-comb. It differs from the Polish comb (page 7) in that the two horns arise from the sides of a median sweUing (Fig. B), In England the Houdan is cultivated with a leaf comb consisting of two broad, flat expansions of the horns arising from a median ridge like " a butterfly with open wings." t It thus resembles the posterior part of a Y-comb (fig. 8). The single comb of the Leghorn is very large and lops in the female to the right or left side of the head. * Petersen, C. E., 1905, p. 961, quoting P. Megnin : "El^vage et engraissement des volailles." t Hurst, C. C, 1905, p. 132. 20 INHERITANCE IN POULTRY. 5. Whiskers, or Muff. — This is a bunch of long feathers growing from the sub-orbital and post-orbital region of the head. This characteristic, of whose origin nothing is known, has been engrafted on several of the other French breeds : the Crevecoeur, the FaveroUe, the Du Mante, the Cossack, the Bourbourg, etc. The muff occurs also on breeds which have little in common with the French fowl, e. g., the Sultan and the Orlaff and Pavaloff of Russia. 6. Beard. — This consists of a number of long feathers growing from the middle of the chin and upper throat region (fig. 16). There is a fold of skin here from which the feathers arise. Such a beard is usually associated wuth the muff. The fold of skin, " dewlap," is found in some Indian Games and, less marked, in the Dark Brahma male. 7. 8. Crest ; Cerebral Hernia. — These are indistinguishable from those of the Polish (pp. 8-10). 9. Foot Color. — The brilliant yellow color of the tarsus of the Leghorn is strikingly different from the colorless or dirty gray tarsus of the Houdan. There must be a special yellow pigment in the skin of the former which is absent in the latter. 10. Number of Toes. — The possession of an extra toe (fig. 13) is an ancient characteristic of poultry. The Latin author Columella, speaking of the fowl kept by the Romans, says : * ' Those hens are reckoned of the purest breed which are 5-clawed, but so placed that no cross-spurs arise from the legs." Since the tendency to extra toes must have arisen spontaneously once, there is always a possibility that it has done so several times, and it is b}' no means certain, although probable, that the extra toe of the Polish is genet- ically connected with that of the Roman fowl referred to. The following record of occurrence of extra toes in poultry is interesting, since in this case no relation to the Roman fowl is probable. A writer * in Der Zoologische Garten states that Carl Bock in his " Reich des weissen Elefant," p. 267, re- lates that he, in a journey from Tschengmai, on the third day reached Muaug Hawt, a way station on the road to Mulmen. This village is dis- tinguished for its 6-toed fowl. Again, the Silky fowl, which certainly have little in common with the Dorking, have extra toes (page 59) . The extra toe is to be regarded as a sport which has appeared at different times and which is easily maintained as a racial characteristic. The question of the inher- itance of such a sport is always interesting. The Houdan has typically only one extra toe, making 5 in all ; it is occasionally found wnth 6. Bateson and Saunders (1902, p. 98) sometimes got 6 toes in hybrids between Leghorns and Dorkings. The length of the extra toe and the completeness of bifurcation are very variable. *Laiigkavel, B. 18S6, p. 35. SINGLE- COMB WHITE LEGHORN AND HOUDAN. 21 PREVIOUS INVESTIGATIONS. During the progress of my experiments appeared the second report to the Evolution Committee of the Royal Society by Bateson and others. This contains a paper by Hurst (1905, p. 133) giving his results with White lyCghorn male X Houdan female. These will be considered in comparison with my results, MATERIAIv. The mothers were two Houdan hens* (fig. 16) purchased from a dealer as pure stock. They agreed well with the standard requirements. When bred with a Houdan male they produced only typical Houdans. Tho. father was a Single-comb White Leghorn f likewise of unknown ancestry. The plumage of No. 13A is devoid of black pigment, and mated with White L,egliorn hens it has produced only White lyCghoru offspring. RESULTS. I. General Plumage Color. — First hybrid generation. Of 41 individuals all were white in plumage (fig. 17), but almost without exception both in down plumage and that of the adult there were traces of black on one or more feathers, particularly those of the back ; more especially was this true of the females than of the males. Hurst (1905, p. 133) got II black chicks out of 105 offspring and in the first plumage these developed into 6 black (all pullets) and 5 barred (all cockerels). Here also females have more pigment than males. Of the white chicks all except two developed black ticking. Doubtless these two were males. Second hybrid goie ration. When these hybrids were crossed inter se, out of 50 individuals 9 were markedly pigmented like the Houdan. This is 18 per cent of all cases, expectation being 25 per cent. With larger numbers Hurst (1905, p. 138) got 24.3 per cent black. Equall)' striking is the occurrence of many pure white individuals along with the impure whites. The pure whites that were reared to maturity proved to be males ; the impure whites were females. First hybrid ( No. Sy $) crossed ivith white Leghorn {No. /z 9 ). The father was pure white ; the mother was speckled with black. Of 22 offspring all were white, but some had single pigmented feathers. All males (9) are pure white, excepting No. 562, which has some black on two feathers of the left wing coverts, and No. 649, which has one-half of one right wing covert black. My only certain female has a score of partly black feathers. Hurst (1905, p. 139) obtained 66 clear white chicks and 69 white, ticked with black. I *Nos. 8 aud 11, received July i, 1904, from Geo. C. Ely. t No. 13A., received Sept. 15, 1904, from a farmer. 22 INHERITANCE IN POULTRY judge this equality to indicate a difference of color in sex ; or else the pig- mented individuals are heterozygotes. Possibly the females are the hetero- zygous forms — the males homozygous. 2. Color of Upper Mandible. — This assumes its final condition so late in life that definite statistics will not be given now. First hybrid generation. A few young are recorded as showing yellow. The rest are white ; this is probably the young condition of the light horn of the adult Houdan. Light horn seems dominant. Second hybrid generation. A few cases of black mandible are recorded, even in the young, where pigment is less developed. First hybrid {8y) crossed with white Leghorn. All older chicks have white mandibles ; there are no blacks. 3. Nostrils. — First hybrid generatioti. Of 25 individuals, all but one have a nostril of one-half width or less — thus approaching the white Leghorn type. The exceptional individual (No. 386 ? ) has a typical high nostril and is peculiar in respect to comb also. Only one individual is recorded as having as narrow a nostril as the Minorca. Seco7id hybrid generation. Forty-nine individuals give : Characteristic. Frequency. Per cent. Narrow (24) and intermediate (8) 32 ^7 65.5 34-5 High Total 49 loo.o On the assumption that " narrow and intermediate " includes pure-narrow and heterozygous individuals, while "high" includes recessive, pure-high individuals, we should expect 75 per cent and 25 per cent in the two classes respectively — only an approximate agreement with the actual. First hybrid {No. 87$) crossed with white Leghorn (71^). The father has a "one-fourth" nostril; the mother, of course, a typical "narrow" one. Of 24 individuals 12 are recorded as narrow ; 12 as intermediate of some grade. This gives the ratio 1:1, which we expect, assuming the inter- mediate nostril to be the heterozygous type ; the narrow, the pure type. 4. Comb. — Fii^st hybrid generation. Of 41 individuals 40 have the Y-shaped comb in some form (fig. 17). This comb resembles that of the Minorca X Polish hybrid. There is no case of a single comb in this generation, but there are two cases in which the posterior end of the comb is merely much thickened. On the whole the present series of cleft combs differs from the former in that a smaller proportion of the comb is split — no cases of wholly split or cup combs occur, although in one important case (87 ^ ) two-thirds SINGLE-COMB WHITE LEGHORN AND HOUDAN. 23 of the comb is cleft. A new characteristic of this series of cleft combs is the occasional appearance of a median comb lying between the two wings of the cleft comb — a posteriad continuation of the single part of the comb. This condition appears in three cases (258 $ , 259 ^ , 448 ^ ). It is important for the interpretation of the cleft comb. It gives the posterior part of the hybrid comb the triple condition characteristic of English Houdans, The one case that lacks the Y-shaped comb is No. 386 9 (with high nostrils). She has only a pair of papillae. Hurst (1905, p. 133) got no single comb in 105 offspring. Second hybrid generation. Fifty-five individuals show the following dis- tribution of comb forms : Characteristic. Frequency. Per cent. Actual. Expected. Single Y comb .... or absent Total 17 23 15 30.9 41.8 27.3 a /s 25 (18.75) 50 (56.25) 25 (25.00) 55 loo.o 100 (100.00) The Y comb being the heterozygous form should appear in 50 per cent of the cases ; each of the other forms in 25 per cent. The deviation from expectation is of the same character as in Series I. That the approximation to theory is less close is probably due to the smaller total number. Hurst (1905, p. 138) obtained 56 ordinary single combs in 226, or 24.8 per cent. First hybrid {8y $ ) crossed with white Leghorn (/z $ ). The Y-shaped comb crossed with single gives, in 26 individuals : Characteristic. /. Per cent. Single 15 II 26 57-7 42.3 Cleft, etc. {^see Remarks) Total lOO.O Remarks: Including two cases in which a median ridge runs through the cleft comb. Of these one is a nearly typical pea comb except that the side lobes are higher than the median one. Including, furthermore, one case of an arrow-shaped comb, two parallel V 's occurring in front and behind, respectively, being joined by a median ridge. Including, finally, two cases of cup-comb. Here the expected equality is approached. Hurst (1905, p. 139) obtained 60 ordinary single combs in 135 individuals of this cross, or 44.4 per cent. 5, Face Feathering. — First hybrid ge?ieration. Of 24 recorded cases all show the muff (fig. 17). 24 INHERITANCE IN POULTRY. Second hybrid generation. All individuals (26) whose face feathering was observed are recorded as muffed; concerning a greater number (35) the record is silent. What has become of the expected 25 per cent of muffless individuals? It is possible, but on strict chance hardly probable, that the uiufBess individuals all died young. A decisive answer to our question must await further experimentation. First hybrid {Sy $ ) crossed i^'ith white Leghorn (7/ 9 )• Only one parent is muffed. Muffed and non-muffed offspring occur in approximate equality ; but even in the adult muffing is not full in amount. This cross confirms the result of the first that muffing is dominant, but it is not perfectly so. 6. Beard. — In \^i& first hybrid generation all individuals are bearded. When these hybrids are crossed with the white I,eghorn about half of the offspring are beardless. Beard is dominant. 7. Cerebrai, Hernia. — \\xt\i& first hybrid generation all (24) individuals were without external evidence of the cerebral hernia. In the second hybrid generation out of 45 individuals 11 had the hernia (fig. 14) and 34 had none, or 24.4 per cent and 75.6 per cent respectively. When the hybrid zvas crossed zcith the white Leghorn (71$) no individual with the hernia appeared. The cerebral hernia is a recessive characteristic. However, the height of the frontal dome is variable, even in the pure-bred Houdans, and on at least one occasion the cerebral prominence in an unhatched h5-brid was so high that it was doubtful whether or not it might represent a hernia. Here, as in Series I, plain-headedness, though clearly dominant, is imper- fectly so. 8. Crest. — First hybrid geyieration. Of 25 individuals all are crested. Hurst (1905, p. 134) gets the same result. Second hybrid generatioii. Of 19 indi- viduals 6 are non-crested, or 31.6 per cent, approaching the expected 25 per cent. The remainder are crested. First hybrid {8j $ ) crossed with white Leghorn (7/9)- Of 15 individuals 6, or 40 per cent, are without crest. Expectation is 50 per cent. Crest is clearly dominant, yet in the first hybrid it is never so large as in the Houdan. This fact is, I take it, due to imperfect dominance. It may, however, be associated physiologically with the absence of a cerebral hernia. 9. Foot Color. — In the first hybrid generation this always becomes white in the adult, although sometimes yellow in young birds. In the second genera- tion of hybrids bred inter se or with the White L,eghorn stock, yellow reappears as leg color. Statistics would be misleading on this point, as permanent leg color is reached only on maturity. It may be concluded that white is domi- nant over yellow. SINGI.E-COMB WHITE LEGHORN AND HOUDAN. 25 10. Number op Toes. — First hybrid generation . Among 37 individuals the following distribution appears : Number of toes. /. Per cent. 4-4 4-5 s-s IV 16.2 83.8 Total 37 lOO.O Hurst (190.5, p. 133) got among 105 birds 103 with trace of extra toe (including duplication of nail and hyperphalangia) and two without any such trace. The difference in the proportions of extra and normal toes between Hurst's and my results is partly a matter of classification and perhaps partly due to the real difference in potency of the extra-toe characteristic in the two strains. Second hybrid generation. To learn if the individuals with 4-4 toes were merely imperfect dominants or true recessives I mated two of them (Nos, 84 9 and 869) with their brother, a 4-toed cock (No. 83 ', 1905, from the Cyphers Incubator Compan3% together with two hens (Nos. *Nos. 129, 130, 131, 132, received February, 1905, from Mr. H. B. Kutschbach (fig. 26, pi. ix). 40 INHERITANCE IN POULTRY. 127, 128). Mated with the hens, nine young were produced. Four of these were typical White lyeghorns without black ; three others were white except that black feathers occasionall}- appeared. One (No. 213) was nearly solid black and one (No. 229) was black with nearly every feather barred with white. It is plain that the strain I have has not been wholly purified of black. This is indicated also by the fact that No. 128 has every feather peppered with black — a heterozygous form of coloration. RESULTS. 1 . General Plumage Color. — Of 26 hybrids, 1 1 were pure white or had only a little black ; 7 were black, sometimes with a little white, and 8 were barred black and white (fig. 27). The results confirm the view that White Leghorn Bantan No. 126 (? has black germ cells. The barred condition is unexpected and is probably recessive in the White Leghorns. 2. Earlobp: Color. — In all cases ( 10) of adults but two, the earlobe is red ; in the remaining two some white is mixed with the red. The red earlobe is probably dominant, but imperfectlj^ so. 3. Vulture Hock, — Out of 13 cases n have clearly no vulture hock and two show a slight enlargement of the heel feathers. Vulture hock is prob- ably recessive. 4. Foot Feathering. — Every hybrid is booted, but the booting is less heavy than in the Dark Brahma (fig. 27). Booting may be dominant, but it is not perfectly so. CONCLUSIONS. The male parent is heterozygous and probably contains at least three sorts of gametes, viz, pure black, pure white, and barred, the last being a mosaic* The black of the mother is recessive to all of these. The occurrence of barred mosaic is of interest, but it is of unknown origin. The ancestral red ear-color and the new ' ' booting ' ' are both dominant. Dominance is, however, im- perfect. Series VIII. — White Leghorn Bantam and BufF Cochin Bantam. STATEMENT OF PROBLEM. This series was undertaken to determine the method of inheritance of buff when combined with a white plumage coloration. THK RACES AS A WHOLE. The White Leghorn Bantam has been described at page 39. The Buff Cochin Bantam (fig. 28) is a diminutive Buff Cochin, which resembles in form the Black Cochin (p. 39). Cochins as we know them to-day (the name was formerly applied to a different, now extinct, race) seem to have been imported into this country and also into England from eastern China * Castle and Allen, 1903, p. 606. WHITE LEGHORN AND BUFF COCHIN. 41 in the year 1847. The earliest importations were buff. According to McGrew (1904, p. 526) : la many of these retreats, mi-au, ^ formed first and growth continues ^^ n ^^p' at the base, within a sheath, for a ,; ^ /^3Q shorter orlonger period, depending ^^ q^, Q on the eventual size of the feather. O Q OP6 (D o 'Q (^As PeO V The reason why some feathers, like ^iQ the contour feathers, are short is because growth quickly ceases. The feathers of the hackle, saddle, ^^^ c.-nia^ram of arrangement of the tail feathers, and tail of the male are long be- ^i-^lj, feathers of right anterior row; .4'i-^ '5, feathers ,1 ,1 • J • • it, of left anterior row; .1/1- .1/7 and .I/'i-.V't, feathers of cause the growth period is in them Hght and lell middle rows; /',-nand PV/'e. feathers prolonged. The sickle feathers of of right and left posterior rows ; o. g-., oil gland; ,1 T -u i.-ii • r /)- K, dorso-ventral line ;/?, right : i, left. the Leghorn are still growing for three months after the molting period ; consequently they attain a length of 300 to 400 mm. If the period of drying up of the growth sheath at the base of the sickle feathers could be delayed in the Leghorn for an entire year they would become each a meter long. The reason for the great length of the tail feathers of the Tosa fowl is that they do not cease growing. In this respect they resemble the long hair of Angora guinea pigs, rabbits and cats, and the head hair of man. Morphology of the tail. As just intimated, only certain feathers of the tail of the Tosa fowl grow indefinitely. It is now necessary to describe the structure of the tail. The feathers have the following arrangement : The posterior row {P) consists of broad feathers with rounded ends and constitutes the characteristic " fan " of the tail. The middle row (J/) con- tributes the characteristic long growing feathers, those nearest the median 46 INIIERITANCK IN POULTRY. line being longest. The sickle feathers (5) may belong to either row, so far as the adult position indicates ; but, as growing feathers, they belong physiologically to the middle series. The anterior row (A) is, at the same time, the posterior row of tail coverts. The lateral feathers of this row are the smallest, owing to a late and brief growth. The long tail of the Tosa fowl is thus produced by the prolonged growth period of the middle row of feathers including the sickle, together with the more median feather of the anterior row. Cause of excessive growth of tail ; Cuiuiinghanf s experiments. The cause of this prolonged growth of the median and sickle feathers is the crucial point. The latest student of the subject, Cunningham (1903, p. 232), quotes Mr. John Sparks as stating : "In order to ensure very great length of tail, the cocks ought to be kept on a perch and the tail-feathers should be pulled gently every morning." Cunningham adds: "My own e:cperiments tend to show that this mechanical treatment of the feathers is the whole secret of the mystery. ' ' He describes in great detail how he stroked the tail of one of tv/o cocks daily ; the other not at all. When a feather otopped growing he pulled it out. He concludes (p. 248) : la the cock whose feathers were stimulated by pulling, growth did not go on at a more rapid rate, but continued for a longer time and produced a longer feather. Thus in cock A [not stroked] no growth took place after April i, and the maximum length was 2 feet 4)^ inches; while in cock B [stroked] growth continued till July 13, and the maximum length was 2 feet 9^ inches. Half a page farther on Cunningham sums up thus : The long-tailed cock in its perfection, therefore, is neither a sport nor a breed, but the product of artificial cultivation ; and the excessive growth of the feathers is the result of stimulation applied to the individual. The more important part of the stimulation is not the mere pulling of the feathers, but the extraction of it which causes the growth of its successor. One can not but remark that Cunningham here contradicts himself. After having laboriously pulled the feathers for over a year and found that the feathers are stimulated b^' pulling, he states : " The most important part of the stimulation is not the pulling but the extraction of the feather causing the growth of its successor." Does Cunningham indeed think that, origi- nally, by extraction of a feather its follicle was so stimulated that it there- after produced feathers which neither ceased to grow nor molted and, moreover, so affected the germ plasm as to produce a race with a tendenc}' toward excessive growth of feathers ? Certainly such a conclusion seems past belief. Author's experiments. To see what influence, if any, stroking the tail feathers has upon their growth, I experimented upon two cocks. One (No. 3, "Admiral Togo") was stroked twice daily by passing the feathers of the middle and anterior row between the thumb and forefinger. The TOSA FOWL AND WHITE COCHIN. 47 other (No. 7, " General Oyama ") did uot have its tail stroked. The two birds were treated similarly, except that Admiral Togo was confined to his perch during all but about two to six hours per day, while General Oyama had free run with the hens. The experiment was begun July 20, 1904, when the cocks were 103 days old, and was continued until March, 1905, when " Oyama " died of roup. The relative growth of the corresponding feathers of the two males is shown in a series of curves (text figure D). The full line is the curve of 600 500 4-00 300 aoo 100 600 500 400 300 200 ICO ^ Aug. Sept. Oct. Nov. Dec. Jan, ^ , Aug. Sept. Oct. ^ Nov. ^ Dec. Jan. y. cl /' ,^ N/i| "A yy ^3 5 y ' y y y k^'^' y / / /""*"' 7 • • / 3 7 / 1/ • / / -^ 7- ,./ /^ ■^ 3 / y ,7 y y M^ / ^^ f 1 1 / y- / 7 ^ y\ r y x- 3^ / ^^' 1 \7 ■ 1 1 Fig. D.— Curves of growth of certain tail feathers of Togo (No. 3, full line) and Oyama (No. 7, dotted line). Ml, first right middle tail feather; M.^, M'j, second right and left middle tail feathers; Ms, third right tail feather ; S, right sickle. the stroked feather ; the dotted line that of the unstroked. These curves show several things. First, the average rate of growth of one of the tail feathers in the Tosa fowl is about 3 mm. per day, or an inch a week. Consequently, if growth is uninterrupted and the feather does not break, it should come to be over a meter long by the end of one year. The extremely long feathers — 5 meters or more — on record are acquired by (a) rapid growth, ((5) continuous growth, (c) preservation of the tail from breakage, and {d) long life of the individual. If stroking has any effect it must be by altering one or more of these elements. Second, the curves show fluctuations in the rate of growth due to fluctua- tions in the condition of the fowl. Third, in the case of those feathers that were studied during the greatest 48 INIIKRITANCE IN POULTRY. period, viz, M,, M.,, M'.^, the stroked feather grew more rapidly thau the corresponding unstroked. Fourth, the unstroked feathers of No. 7 ceased growing earlier than the corresponding stroked feathers of No. 3. Whether the third and fourth items are due to differences in treatment or to other peculiarities of the fowl can not be asserted definitely. In any case the feathers all eventually, at about six months, ceased to grow. Owing to the death of No. 7 soon after, the comparison had to be abandoned. The result agrees with Cunningham's in that stroking prolongs the period of growth ; but the result, depending on three feathers, can hardly be generalized. It would not be surprising, in view of what we know of functional hypertrophy, if it were some day demonstrated that stroking always prolongs the growing period of a feather. This is, however, an entirely different matter from proving that the origin of the long-tailed con- dition of the Tosa fowl was due to, and its maintenance in some way depends upon, stroking. A few further cxpeiiuiencs have been made with Admiral Togo (fig. 31). I have found, in confirmation of Cunningham, that if a feather that has ceased to grow be forcibly removed it is quickly replaced by another that continues to grow. Thus a feather pulled out January i, 1905, had grown steadily to November i ; but as the bird was needed for breeding and could not be confined, the tail has repeatedly broken off. In September, 1905, it measured over 900 mm. As a further criterion of the value of manipulation in causing this great growth of the tail feathers of the Tosa fowl, it becomes important to see bow this physiological characteristic is inherited when crossed with a short- tailed individual. This consideration led to the present series of experiments. The tail of the Cochin fowl is the shortest of all races of poultry. Thus Wright (1902, p. 245) expresses the ideal of the fancier : " The tail of the cock should be as short as possible." 3. Foot F'eathering. — While the Cochin is very heavily feathered on the foot, the Tosa fowl is typically clean-legged. However, No. 3, which is not the father of any of my hybrids, shows a few bunches of rudimentary feathers or hairs on the tarsus. 4. Foot Color. — The willow foot of the Tosa fowl is derived directly from the Jungle fowl. The white foot of the White Cochin seems to be an albinic form of the yellow foot derived from its Indian-Malay ancestry. MATERIAL. First Generation. — The mother was a White Cochin Bantam, No. 35a (fig. 32), of unknown origin, but apparently pure in respect to the four characteristics here under consideration. The father was imported from Japan, having been purchased in New York city, January, 1904. It has a dark Game coloration (fig. 29). TOSA FOWL AND WHITE COCHIN. 49 Second Generation. — Two hybrid cocks, Nos. 53 (fig. 34) and 95 (fig. 35), were successively bred to their sisters, Nos, 58 cfig. 33), 94, 96, and 98. RESULTS. I. General Plumage Color. — First hybrid generation. Of 7 offspring, 3 cocks and 3 hens developed their adult plumage. The males were all of the male Tosa-fowl coloration except that every feather was repeatedly barred with white (figs, 34, 35, 37A). The females were all of the female Tosa-fowl coloration except that the shafting was much broadened (fig. 37) ; also the saddle feathers and the proximal secondaries were obscurely barred black- and-buff. Second hybrid generation. Among 57 individuals we have the following distribution of plumage color : Color. No. Per cent. White 16 41 28.1 71.9 Pigmented The original white color has reappeared in about one-fourth of the cases (fig. 38) ; plumage color segregates in the germ cells of the first hybrid generation in true Mendelian fashion. Of the 16 whites, only 5 were without trace of reddish pigment. Such pigment occurred on the breast, top of head, and remiges. The purity of the germ cells from which these whites sprang — the completeness of segregation — is not always perfect. The 41 pigmented individuals show a curiously mixed lot of coloration. Of 14 vsx2X\xxQ. females, 6 are like the female Tosa fowl, without barring, but sometimes with wider shafting than male Tosa fowl. The remainder have feathers of the back and wing coverts barred with lighter, even with white — a condition not found in the female first hybrids. One of these (No. 659) shows a mixture of female Tosa and female Partridge Cochin coloration. As no Partridge Cochin is involved in the immediate ancestry, this looks like a * ' reversion ; ' ' the characteristic has probably lain latent in the White Cochin. Of 10 males, two show no trace of white, and may, consequently, be considered as homozygous. The remainder are more or less barred with white. One bird (No. 646) shows a remarkable mixture of Tosa and male Partridge Cochin coloration. 2. Tail Length. — First hybrid generation. All the three males reared developed abnormally long middle tail feathers. One of these birds died young. The second bird (No. 53, fig. 34) lived to be exactly one year old. Its sickles were 427 mm. long and had stopped growing. It had suffered a severe paralytic stroke four months before its death. The remaining cock (No. 95, fig. 35) had a.t iij4 months sickle feathers 360 mm. long and still growing. These feathers had thus grown at a rate of about i mm. a day, or 4 50 INHERITANCE IN POULTRY. ouly one-third that of its father. The long-tailed characteristic of the male has been inherited, but in a reduced form. Second hybrid generation. Still immature. 3. Foot Feathering. — First hybrid generation. Of the 7 individuals all have the feet feathered (" booted ") and the females are provided with a " vulture hock." The feathering is usually less than in the Cochin. Second hybrid generation. Among the 55 individuals of this generation all degrees of foot feathering were obtained. Eight cases are recorded as " heavily booted," 27 as "booted," 13 as slightly booted, and 7 as non- booted. The classification is arbitrary and therefore the exact proportions not significant. The important outcome is that a good share of this genera- tion is essentially clean-legged like the Tosa-fowl ancestor, and an approxi- mately equal proportion is heavily booted like the Cochin ancestor, while the rest are feathered to an intermediate degree like the parents. 4. Foot Color. — This has a curious way of changing during the early months of the individual. White is often represented in the young by yellow. A ' ' slate blue " or " bluish black ' ' occurs ; this may be a form of the willow from which yellow has been extracted. First hybrid generation. Of 5 individuals two are recorded as white, one as yellow, one as willow, and one as slate blue. Here is practically equal frequency of the light and dark types. Second hybrid generation. Fifty-three individuals give the following dis- tribution of foot color : Color. /. Per cent. White Yellow II 16 20 6 Willow Slate or bluish Total 53 100 This shows a practical equivalence of light and dark foot colors as in the first generation. The interpretation of this result must be left for later studies. 5. Correlation of Characteristics. — Considering only the three characteristics of plumage color, booting, and foot color, and assuming that game color and boot are dominant and light and dark feet equally apt to occur, we find the following calculated and actual frequency of each combi- nation (actual percentage is in italics) : Game plumage . White plumage.. , To;* 69.Si- Booted 5C.3^ 58.5';, Non-booted IS.TfS (Booted 18.75^ Non-booted 0.25$ ii.4^ Li^htfeet 28.1^ Dark feet 23.15^ 'Light feet 9.4^ Dark leet 9.' OS ,^ f Light feet 9.45? ■"*•*'" I Dark feet 9.4^ ,„y J Light feet 3.1^ ^•''>" t Dark feet S.ljf 50.8^ t8.Si, 5.7^ 5.7i 15. li. O.Oji 1.9'^ DARK BRAHMA AND TOSA FOWI,. " 51 Consideriug the intrinsic diificulties of classification due to the partial blending of characteristics, there is a fairly close correspondence between the calculated and the actual. This result prov^es that there is little if any necessary correlation between the characteristics in question ; they may combine in a chance fashion in the second hybrid generation. CONCLUSIONS. The inheritance of color in this cross between a white and a game-colored breed is remarkable in that white is not dominant — as is usually the case — nor recessive ; but inheritance is particulate in the heterozygote, producing barred offspring. Segregation nevertheless occurs in the second hybrid generation, but the extracted whites and game colored birds are, for the most part, no longer as pure in color as their grandparents were. The germ cells are no longer perfectly pure — they have become infected by con- tact with the opposite quality. The long-tailed characteristic behaves in inheritance like a unit character — in no wise different from plumage color. One can not help doubting whether it originated by any different method from that in which the diverse colors of poultry have arisen. Foot feathering is dominant here as in many other cases ; yet the domi- nance is incomplete. The germ cells of the second hybrid generation are no longer pure. The White Cochin has no sexual dimorphism in plumage color, while the Tosa fowl is strongly dimorphic. Every one of the first hybrids is dimorphic in plumage coloration, the two sexes resembling, except for the white, respectively the female and the male Tosa fowl It is striking to see how from a germ cell of the male Tosa fowl either a bird colored like a male Tosa or a bird colored like a female Tosa may arise. The male germ cells contain the Aniagen not only of the male characteristic but also of the female characteristic (Darwin, 1876, Chapter XIII). Series X.— Dark Brahma and Tosa Fowl. STATEMENT OF PROBLEM. This series was undertaken primarily to test inheritance of secondary sexual characteristics and the possibilitj- of transferring them from one sex to another. the; races as a whole. The Dark Brahma male and female have been described at page 32 ; the Tosa fowl, male and female, at pages 43, 44. Each race has a strongly marked sexual dimorphism in plumage color. The males have feathers of a more uniform color ; the female Dark Brahma has penciled feathers ; the female Tosa fowl has mossy feathers with prominent light shafting. 52 INHERITANCE IN POULTRY. TABLE OF CHARACTERISTICS. No. I 2 3 4 5 6 7 8 9 lo II 12 13 Tosa fowl. Characteristic. I Female. vShafting Lacing on hackle. Lacing elsewhere. Penciling Red wing-bar. . . . White wing bow . Comb Earlobe Iris color Foot color Vulture hock Foot feathering.. . Tail feathers . . . Male. See page- Present . . Absent. . Present . . ! Present . Absent. . . Absent. . Ab.sent . . Absent. Absent. . . Present . Absent. . . Absent. . Single While, red edge. Red Willow Absent Absent , Long 32 33 33 48 48 44-48 Dark Brahma. Female. Male. Absent . Present Absent. Present Absent. Absent. Pea Red or bay Yellow. . . Yellow . . . Present . . . Present . . . Short Absent. Present Present Absent. Present Present See page- 33 33 33 34 34 REMARKS ON THE CHARACTERISTICS. 1 . Sh.\fting. — In plumage, shafting is a light streak on the shaft and adja- cent parts of the vane. Of the two parental races it occurs only in the female Tosa fowl (fig. 30). The light shaft-stripe is, however, widespread among females of certain dark or silvered races — e. g. , Silver Wj'andotles, Silver-gray and Dark Dorkings, Silver Duckwing Games, and Silver Penciled Hamburghs. It crops out in many individuals where its occurrence is regarded by the " fancy " as a " defect." It is an original characteristic of poultry introduced from Gallus ba?ikiva, whose female exhibits it conspicuously (fig- 39)- 2. Hackle Lacing. — Among most broken- colored poultr)' the hackle feathers are unlike those of the rest of the plumage. Usually the hackle has a dark band in the center and is margined or laced by white — more rarely by yellow or red. In the female Jungle fowl (fig. 39) the feathers of nape and hackle hav^e a black center (with broad, straw-colored shafting) and are laced with straw color. The male Jungle fowl has hackle feathers of a solid red color. In the descent of the domestic poultry, hackle lacing seems to have been transferred to the male sex also. 3. Body Lacing. — Few races of poultry exhibit lacing elsewhere than on the hackles. It is very prominent on the Indian female, but is not found on the Jungle fowl of either sex. In the Dark Brahma male (fig. 19) it occurs only on the saddle feathers. Whether its laced saddle is derived from the Indian or is due to a spreading, through correlation, from the hackles can not be said. Lacing is found on the breast of Game fowl and over much of the body of the female Dark Dorking. Among certain derived races, such as the Spangled Polish and the Laced Wyandottes, it affects nearly the whole plumage and is ver}- conspicuous. DARK BRAHMA AND TOSA FOWL. 53 4. PenciIvING. — This may be defined as a concentric repetition on the feather of alternating bands of the lacing and the ground color. In the hackle of the female Jungle fowl the straw color of the lacing is repeated in the center, the two light areas being separated by a black band. In the female Indian fowl the feathers of the throat are laced, but lower down on the larger back- feathers and on the wing bows, there is a second or inner lacing — /. e. , the wing is penciled ; * consequently penciling may be said to be a fundamental form of coloration in the genus Callus. Penciling occurs widespread among the derived or secondary races of poultry, particularly in the " partridge " varieties. A curious modification of penciling is the straight transverse barring of the feather familiar in the Barred Plymouth Rock and Penciled Hamburghs. 5. Red Wing-Bar. — The wing-bar is formed by the lower wing coverts, usually the first to third rows above the remiges or flight feathers. In the male of many races of fowl these differ from the more proximal rows. In the Dark Brahma male they have white and red in addition to black. The wing-bar has probably been derived by the Dark Brahma male from the Indian fowl. In the male Tosa fowl the lower wing coverts are tipped with red, but they show no white. 6. White Wing-Bows. — The wing-bow is formed by the upper or proximal rows of wing coverts — /. e., above the third. These coverts are frequently of a different color from the wing bar. They are red in the male Indian and Malay, but they are white f in the male Aseel.| The white wing-bow of the Dark Brahma has probably been derived from this source. The wing- bow of the male Tosa fowl, like that of the Jungle fowl and Games, is red. 8. White Eari,obe is a derived color, the primitive condition being red (page 33). 9. Iris Color. — The origin of the yellow eye of the Brahma has been dis- cussed at page 33. The red eye of the Tosa fowl is found in most Games and is the prevailing color among domestic poultry. MATERIAL. Mother. ~'^o. 121, Dark Brahma Bantam (fig. 18). § She is a beautifully penciled bird, with horn-colored beak, pearl-colored iris, prominent vulture hocks, and booted down to the outer two toes. To test her purity, she was bred for a month to No. 122, Dark Brahma male, also from Mr. Hodges. Their offspring died before hatching except one (No. 146 ^), which is a typical Dark Brahma. *CompareWriglit, 1 902. p. 334, and American Standard of Perfection, 1905, p. 207, figure. t According to Ludlow's painting in Wright, 1902, opposite p. 326. X Since the above was written I have purchased a male Aseel which has dark coverts tipped with white. § Weight 1,300 grams, received February, 1905, from Mr. F. H. Hodges Red Bank, New Jersey, marked F. H. H., No. 66, also No. 338. 54 INHERITANCE IN POULTRY. Father. — A Tosa fowl bred at the station, No. 8a, "General Oyama," referred to at page 46. RESULTS. The produce was 5 females and 16 males (fig. 40). They are all blocky birds, verj- different from the Tosa fowl, but longer than the Brahma. The maternal or Brahma type is, however, predominant. Only the first genera- tion of hybrids has been reared. 1. Shafting. — The male hybrids are mostly without shafting on the feath- ers of the back and the wing coverts. Two, however, show clear yellow shafting on these feathers, and in two others the shafting is a light buff color. The female hybrids have these feathers shafted. Shafting is domi- nant in the female hybrids. It is doubtfully transferred to some males. 2. Hackle Lacing. — This showed on all hybrids of both sexes. 3. Body Lacing. — In the male hybrids the saddle feathers and sometimes the tail coverts are laced with vellow as in the Brahma. Such lacing does not appear on the female. Lacing in the male sex appears to be dominant. 4. Penciling. — This appears as typical penciling or as barring on the back and saddle and on the exposed parts of the secondaries of the female hybrids. It does not appear on the males. Penciling seems to be dominant over mossiness and to be confined to the female sex. 5. Red Wing-Bar. — This is present in all of the first hybrid males, but the red is deeper and spreads farther over each feather than in the Dark Brahma, the red of the Tosa fowl having its effect. The female is without wing-bar as in the female parents. 6. White Wing-Bow. — Of 13 hybrid males four show no white in the upper wing coverts (fig. 40) ; but one of these has a light buff bow — a tendency toward white. The others have a small amount of white, which is derived from the Dark Brahma. The white has, however, been clearly reduced in amount. The interpretation of this result must await further breeding. 7. Comb. — In every hj^brid the comb is pea, proving the dominance of that form over the single. The pea is, however, often atypical, the lateral ridges being rudimentary. Dominance is not always perfect. 8. Earlobe Color. — Every hybrid shows some white, as in the Tosa fowl ; but this white tends toward yellow — a much diluted red. White seems to dominate, but, if so, the dominance is imperfect. 9. Iris Color. — This is red in the hybrids; but in two cases the red ap- proaches orange. The iris color of the Tosa fowl is dominant, but imper- fectly so. 10. Foot Color. — Of 21 hybrids, all males (16) show yellow feet and all females (5) willow feet. This dimorphism is not found in the parent races. 1 1 . Vulture Hock. — The hybrids show a tendenc}^ toward long feathers hanging over the heel (fig. 40). In one case these had reached a length of 105 mm. by six months ; in another, about 90 mm. In other cases these FRIZZLE AND SILKY. 55 feathers are much reduced from the Brahma type, and in one or two cases it is doubtful if they are present. We have to do here either with a blending characteristic or else a very imperfect dominance of the vulture hock. 12. Foot Feathering. — This is always present in the hybrids, but is usually less heavy than in the Dark Brahma (fig. 40). Booting is dominant, but is imperfectly so. 13. Tail Feathers. — 'As none of the hybrids are over six months old, it is impossible to report fully on the inheritance of this characteristic. While in some male hybrids the tail feathers already surpass in length the middle tail feathers of the adult Brahma parent and are still growing, in no ca.se have they made the extraordinary growth of the Tosa fowl. CONCLUSIONS. Method of Inheritance. — The color characteristic of shafting and penciling in the female, and body lacing, red wing bar, and white wing bow in the male, appear to dominate in the respective sexes ; but dominance, if such it is, is alwa3'S imperfect, in that traces of the opposite allelomorph are sometimes found. Furthermore : Red eye color dominates over yellow (not always perfectly). Booting is dominant over clean leg. Earlobe color is something of a mixture. Vulture hock is sometimes very imperfectly " dominant." The length of tail feathers is perhaps a blend. Sex in Inheritance. — For the most part a sexually dimorphic charac- teristic is inherited only by the proper sex. In the hybrids of this series, however, shafting seems to have been partially transferred from the female to some males. Most peculiar is the inheritance of foot color, where all the female hybrids show the willow foot of their father, and all male hybrids the yellow foot of their mother. Series XI.— Frizzle and Silky. STATEMENT OF PROBLEM. This series of crossings was made to learn the inheritance of the allelo- morphs given below. THE RACES AS A WHOLE. The origin of the Frizzle fowl (figs. 41 and 42) is not definitely known. Darwin (1876, Chapter VII) states that they are not uncommon in India, and Temminck states that they are domesticated also in Java, Sumatra, and all the Philippine Islands, being prevailingly white. They must have been brought to Europe early, since they are described by Aldrovandus in 1645 from a specimen sent him from Parma. Willoughby, in his Ornithology (1676), says that he had seen them in England. The recurving of feathers is found in many species of birds. It usually occurs on the neck, where it forms a ruff ; more rarely over the entire body. Frizzled canary birds are 56 INHERITANCE IN POUETRY. occasioually exhibited. Frizzling is probablj' morphologically related to " rough coat " iu mammals. The frizzled characteristic is a typical sport. The Silky fowl (fig. 43) is likewise of great antiquity. Marco Polo saw it in Asia in the thirteenth century (teste, Diirigen, 1886, p. 298). Gesner described it in 1555. It is a native of eastern India, coming, according to Blyth (Tegetmeier, 1867, p. 221), from China, Malacca, and Singapore. A condition allied to silkiness (described below at page 57) is found in other races of poultry, particularly, as the following statements show, iu the Cochins. Tegetmeier (1867, p. 46) says : The singular variety known as Silky Cochins, or sometimes as Emu fowls, is simply an accidental variation of plumage which occasionally occurs and which may be perpetu- ated by careful breeding. The cause of the coarse fluffy appearance of these remarkable fowls is to be discovered in the fact that the barbs of the feathers instead of being held together by a series of hooked barbules (so as to constitute a plane surface, as occurs in all ordinary feathers) are perfectly distinct, and this occasions the loose fibrous silky appearance from which the fowl obtains its name. An engraving of such a feather is given by that author at page 224. Wright (1902, p. 255) states that he has seen no Emu fowl " now^ for twenty years," and makes the suggestion that this entire "silkiness" of feather is the extreme limit, perhaps, of the kind of plumage which gives fluffiness to the leg region of American Buff Cochins. The fluff of Cochins and Brahmas has indeed many points of similarity in structure with the feathers of the Silky. In one feather from the abdomen of a Brahma hen, whose shaft is 35 mm. long, I find the barbs very long (up to 30 mm.) and not connected together. Each barb bears, proximally, two rows of short, flat, hook-shaped barbules alike on the two sides. Beyond, there are a few short barbules that taper to a hair-like apex. Still more distall}' on the barb the barbules may attain a length of 5 mm. , be altogether devoid of booklets, but show a segmented condition as in the Silky Far from my preconceived notion, I find few intergrades between the short barb- ules and the others. The more proximal of the long barbules are the longest of all, and the short barbules (which rarely exceed 0.5 mm. in length) also occur here scattered among the long ones. There thus seems to be a dis- continuity between the two kinds of barbules, and this harmonizes with the view that the long barbule is a mutational form of the more typical short barb tile. As to the relation of the plumage of the Silky fowl to the fluff of Cochins, I have formulated the following hypothesis : Long and short barbules are two dimorphic forms found among birds. This dimorphism has been recog- nized in the terminology "down feathers" and "contour" + "quill" feathers. Down feathers may or may not have a shaft ; they have barbs, and usually barbules, the latter being long and devoid of cilia or booklets. In the contour and quill feathers of most birds the short barbules alone are FRIZZLE AND SILKY. 57 present. But in some birds the barbules are long and devoid of cilia or hook- lets as in the Ratitae (ostrich, emu, cassowary, etc.). In poultry the down feathers are characterized by absence of booklets, and the ventral abdominal feathers of poultry belong to this category. In the Silky fowl the contour feathers, in the strict sense, are absent, or rather they have gained long bookless barbs, and consequently liave become in so far down feathers. But the feathers of the Silky fowl have one new characteristic not found in any other long-barbed forms, namely, the bifurcation and anastomosis of the barbs (page 58). TABLE OF CHARACTERISTICS. No. Characteristic. Frizzle fowl (Game). Silky fowl. I 2 3 4 5 6 7 8 Plumage color Comb form Shaft of contour feather. Barb length Barb form Number of toes Skin color Dark, black, red, and buff. Rose Recurved Short Twisted about long axis. . . Four White White. Single. Straight. Long. Straight. Five or six. Black. Present. Crest Absent REMARKS ON THE CHARACTERISTICS. 1 . Plumage Color. — This characteristic is very variable in Frizzles, owing to the fact that fanciers have established no color " varieties," although an effort is now being made in that direction.* As stated below in detail, my Frizzles were of varied and mixed colors. 2. Comb Forms. — The "American Standard " calls for single comb in the Frizzle and rose, or rather strawberry, comb in the Silkies. My Frizzles have, on the contrary, a rose comb, and my Silkies either a .single comb or a rose comb, the Silky being impure in respect to this characteristic. 3-5. Feather Form. — In the Frizzles the contour feathers have the shaft curved so that its outer surface becomes concave. This is most striking in the neck region, where a ruff is formed (fig. 41). The wing primaries are modified in another direction, since in them the barbs, in groups of 4 to 8. are twisted in corkscrew fashion about their own axis and through 180° or more ; consequently the gray surface, which is normall)- next the body, comes to lie outermost. Such a twisting of the barbs sometimes occurs in primaries of non-frizzled races ; particulary I have found it in the eighth primary of some Houdans. The barbs of the remiges of the Frizzles are mostly .short, and in some cases are lacking altogether, being easih' broken off. The feathers of the Silky fowl are remarkable in all parts of the plumage. The contour feathers, as already stated, are down feathers, whose shaft is *The new "American Standard of Perfection," published by the American Poultry Association, 1905, p. 248, directs that color should be " solid — black, white, red, and bay admissible, provided the birds match when shown in pairs, trios, and pens." 58 INHERITANCE IN POULTRY. usually delicate but not othenvise atypical. The barbs are, on the other hand, remarkably long. Thus in a contour feather, from the middle of the dorsal region, whose shaft is 25 mm. long, the prevailing length of barb is 35 to 45 mm. The barbs are, moreover, remarkable in that they frequently bifurcate, even repeatedly. In a feather before me, one barb, taken at random, undergoes bifurcation four times. As the branches are not all in one plane, the feather becomes exceedingly fluffy. At the proximal end of the shaft the barbs arise parallel and produce an imperfect web close to the shaft, but marginally the web is lost. Distally on the shaft the barbs arise more irregularly from the shaft, often bifurcating almost immediately, so that no web or vane is formed. The barbs may also anastomose. The barbules are not less strikingly modified than the barbs. They attain a length of from i to 2 mm. Moreover, it is not possible here, as in other races, to distinguish between a distal series of barbules carrying a row of booklets or cilia and a proximal series without booklets but with a folded edge into which the booklets of the distal barbules catch. This impossibility is due, first, to the fact that the barbules are not in two series merely, but may arise in three planes, or irregularly ; also, morphologically, all the barbules on the barb are alike. They are all segmented like the ordinary distal barbule, and the booklets are represented by minute thickenings at the end of each segment. As a consequence of this structure the barbs do not hang together to form a vane and the fluffiness is still further exaggerated. The quill feathers of the wing (remiges) and tail (rectrices) of the Silky are modified, but to a less degree. Primaries, secondaries, and coverts are all affected. The proximal part of the vane is nearly normal ; the distal part has barbs of t\vice to thrice the normal length. The barbs may bifurcate repeatedly and even anastomose in the plane of the vane. The barbules also are modified, being much shortened. Proximal as well as distal barbules may carry booklets, as is seen in the middle part of the feather. In the proximal part of the feather, on the other hand, the proximal barbules are without booklets. The feathers of the tail have the web even more broken up than those of the wing. The silky condition of the feather is a characteristic that is either entirely new (progressive in de Vries's sense) or possibly latent (in de Vries's sense) in typical fowl, so that its appearance in the Silky is a case of ' ' degression ' ' (de Vries). If the former, we should expect, according to de Vries, the offspring between a Silky and a non-Silky to show a mosaic of the parental feather characteristics and a non-Mendelian inheritance of silkiness ; if the latter, a recessiveness of the varietal characteristic of silkiness and its Men- delian inheritance.* * De Vries, 1905, p. 280 : " The character of the species is dominaut iu the hybrid, while that of the variety is recessive." On the latter of the two assumptions made above, plain plumage is the species character ; silky plumage, the varietal. FRIZZLE AND SILKY. 59 6. Number of Toes. — This is constantl}' four in pure-bred Frizzles. In Silkies a fifth toe is always present. The extra toe frequently has a double nail, or the division may be complete, resulting in six toes. 7. Skin Color.— In the case of the Frizzle the skin is white, sometimes tinged with yellow pigment. The skin of the Silky is notoriously blue-black. This is a clear case of melanism, and since early times has been associated with the other peculiarities of the Silky. The melanic condition affects the periosteum also. It is remarkable that despite this excess of pigment ren- dering black the internal tissues, skin, leg scutes, comb, and wattles, the plumage should be always white. MATERIAL. Mothers. — Four Frizzles (Nos. 14A, i8a, 19A, and 20A), hatched May, 1904, from eggs obtained from Dr. A. G. Phelps, of Glen Falls, N. Y. All have rose combs and slightly booted feet. No. i8a is peculiar in that the feathers on head and neck are sparse and small (fig. 42). In general color the hens vary ; i4Ais prevaihngly dark brown ; i8a is yellowish ; 19A is light brown, and 20A is mixed black, yellow, and red. A male Frizzle from the same lot of eggs was highly colored red and black. Father. — A white Silky cock (No. 24A, fig. 43), likewise hatched from eggs sent in May, 1904, by Dr. Phelps. The Silky cock and Frizzle hens were mated from January 16 to April 14, 1905. Trap nests were not used, so that I could distinguish mothers only by the form of the eggs. The ^'g% of i8a was very peculiar and was early identified. A certain proportion of the offspring can not be assigned to any particular mother. RESULTS. Only the first hybrid generation has been obtained. 1. Plumage Color. — Of 32 hybrids, 7 (22.6 percent) are white (showing some buff in six cases) and 25 (77.4 per cent) are dark. No. i8a appar- ently produced only dark birds, largely dead-black. The others produced in part white hybrids (fig. 44), but mostly pigmented ones. The result is not what we should have expected. If white were recessive, o to 50 per cent, if dominant, 100 per cent, of the offspring should be white. Moreover, the Silky is doubtless homozygous in respect to color, since (i) Silkj- fowls are carefully bred for white color, and (2), bred to a hen of its own strain, it has produced onl}^ white birds. I conclude, therefore, that the white plumage color is not always dominant over the black, red, and yellow of the Frizzle. The matter will be further investigated. 2. Comb. — In all cases the rose comb of the Frizzle dominated over the single comb of the Silky (fig. 44). 3-5. Curving of Shaft, Barb Length, and Barb Form. — These are all correlated in the first generation. Of 10 mature birds, 6 are typically frizzled 6o INHERITANCE IN POULTRY. and 4 have flat feathers. Assuming frizzling to be dominant, non-frizzling recessive, and that all my Frizzle fowls are heterozygous, we should expect 50 per cent frizzled offspring. The result accords well with these hypotheses. None of the hybrids show any trace of silkine.ss. Silkiness is recessive as against non-silkiness. This result is striking and has been observed by others. Tegetmeier (1867, p. 224) bred Silkies to other varieties and found that " the chickens produced seldom had the silky feathers, but were clothed in plumage of the ordinary character." Mating these hybrids together he got among plain feathered offspring " one covered with feathers like those of the Silk fowl," but with black plumage. Darwin (1876, Chapter VII) had previously bred a white Silk hen to a Spanish cock ; ' ' none inherited the so-called silky feathers." 6. Number of Toes. — Thirty hybrids gave the following distribution of characteristics : Characteristic. /• Per cent. 4 toes, both feet 4 and 5 toes 7 9 14 30 23-3 30.0 46.7 5 toes, both feet Total 1 00.0 Here, as elsewhere in this paper, the inheritance of extra toe is difficult to account for on the Mendelian principle of dominance. 7. Skin Color. — All hybrids have a black skin, Tegetmeier (1867, p. 224) got the same result. 8. Crest. — So far as noted, all mature hybrids have a well-marked crest, but it is somewhat smaller than that of the Silky. CONCLUSIONS. A final conclusion as to dominance must await the production of the second generation of hybrids. The following (in italics) appear to show Mendelian dominance over the corresponding allelomorphs : Rose comb vs. vSingle comb. Frizzle feathers vs. Plain feathers. Black skin vs. White skin. Crest vs. Plain head. Plumage color and number of toes are unit characters, but behave pe- culiarly. The dominance of the crest is imperfect. LEGHORN AND RUMPLESS GAME. 6l Series XII.— Single-comb White Leghorn Bantam and Black-breasted Red Rumpless Game. STATEMENT OF PROBLEM. This cross was undertaken primarily to test the inheritance of rumplessuess, and secondarily of the more primitive game coloration against white plumage color. THE RACES AS A WHOLE. The White I^eghorns have been described at pages 37 and 39. The Black- breasted Red Game closely resembles the wild Jungle fowl in color (figs. 45 and 46). TABLE OF CHARACTERISTICS. REMARKS ON THE CHARACTERISTICS. Uropygium. — The absence of uropygium is a characteristic that has long been known among fowl, but there seems to be little knowledge of its iror- phology. In ordinary fowl there are five free caudal vertebrae, followed by a fused portion — the uropygial bone. In the case of a rumpless Game female (No. 119, fig. 45) dissected by me, there are two unsymmetrically formed and intimately fused vertebrae behind the fifteenth synsacral — the posterior limit of the sacral vertebrae. That there are two is shown by distinct trans- verse processes with spaces of the passage of the nerves. Behind these is a knob of bone about i mm. in diameter. These three elements constitute the entire caudal skeleton. It is profoundly reduced from the normal. Rumplessuess may be found in any race. It has cropped out in two of the 800 fowl bred at this station in the past year — hybrids derived from the Minorca-Polish and the Leghorn-Houdan crosses. It seems like a misuse of the term breed to speak of a " Rumpless breed," as poultry books do. The characteristic is referred to by Aldrovandus in 1645, by Temminck, and by other early writers. Its origin has been ascribed to Persia, to Cey- lon, and to China ; doubtless it occurs in all these places as well as in many others, Taillessness early appeared among fowls in America. Clayton (1693, p. 992) asserted that he had observed that in "Virginia" most of the cocks and hens were without tails, and Wright states that he was informed by a West Indian in 1872 "that the greater number of fowls in his own neighborhood had no tails." Darwin (1S76, Chapter VIIj refers to this characteristic and states that one bird he examined had no oil gland; the same is true of the three rumpless Games that I have had. Among the 62 INHERITANCE IN POULTRY. poultry books that describe the "breed " quite fully are Tegetmeier (1867, pp. 230-232), Baldamus (1896, pp. 170-172, " Kaul oder Kliitthiihuer " ), Diirigen (1886, pp. 98-100), Wright (1902, p. 481), and Weir-Johnson- Bruwu (1905, pp. 1016-1017). Regarding the inheritance of this characteristic, statements are not in accord. Tegetmeier (p. 231) says: A frieud of mine purchased a successful pen [of Rumpless fowl] at a poultry show, taking them away to a walk where no other fowls ever trespassed, and yet the chickens were, in a considerable number of instances, furnished with fully developed tail feathers, being not rnmpless. On inquiry of the previous owner, he stated: " Mine have always done so from the first time I kept them; but the tailed birds will very probably produce rumpless chickens." Three such birds were purposely retained, and they produced the next year more than twenty youngsters, all of which but one were rumpless and destitute of tail feathers. The foregoing experiment would seem to prove that the rumpless parents were heterogametous, and that while rumplessness is dominant the recessive conditionof tail is here prepotent (Castle, 1905). Darwin (1876, Chapter VII) possessed a rumpless bird which " came from a family where, as I was told, the breed had kept true for twenty years; but" he adds, " rumpless fowls often produce chickens with tails." The breeding true of a character may mean either that it is dominant and homogametoits in this respect or that it is recessive. Diirigen (1886, p. 99) states that a rumpless cock mated with a tailed hen produces not exclusively rumple.ss, but a fair percentage of them, and Wright (1902, p. 481) saj'S that " a Rumpless fowl crossed with any other generally produces a large majority of Rumpless birds." All of the fore- going results are consonant with the conclusion that rumplessness is typically dominant, but that the recessive full tail may be prepotent. MATERIAL. The mother was the White Leghorn bantam No. 127 discussed at page 39. She is heterozj^gous and contains black gametes. The father (No. 117, fig. 46) was one of three rumpless bantams obtained from Dr. A. H. Phelps, of Glen Falls, New York. Two of these were typical Black-breasted Red Games; they lack oil glands and weigh about 1,000 grams each. RESULTS. Only the first generation of hybrids has been so far obtained. General Plumage Color. — Of 24 hybrids 12 were white or prevailingly so (fig. 47). Usually, however, more or less black and more rarely some buff was present. The other 12 were either black-and-white barred (and these were all males) or black with more or less reddish. As we have seen, the white mother contains recessive black or black-and-white, so that the result accords with the expectation of only 50 per cent white. BIwACK COCHIN AND RUMPI.ESS GAME. 63 Beak Color. — In the hybrids the beak is sometimes yellow, sometimes black, sometimes black-and-yellovv streaked. Uropygium. — Of 24 hybrids the uropygium is normal in 23 (fig. 47). One chick taken from the egg is recorded as without tail, though tail gland is present. It is doubtful if much stress may be laid on this record, as the uropygium is always ver}' small in the unhatched bird. We may exclude it from present consideration. This whole result was unexpected because opposed to the earlier observations. It leads to the provisional hypothesis that rumplessness is recessive in my strain. If full tail is recessive, then in my strain the recessive condition is prepotent. Further discussion must be deferred until the second hybrids have been bred. Foot Color. — This was 3'ellow in about half of the cases and willow or dark in the other half. Recalling that the White Leghorn is heterozygous, the result favors the hypothesis that yellow is dominant over willow. CONCLUSIONS. White plumage color seems to be dominant over game color. The hy- potheses seem to be warranted that yellow beak and foot color are dominant, and that rumplessness is recessive in this strain. Series XIII,— Black Cochin Bantam and Black Breasted Red Rumpless Game. STATEMENT OF PROBLEM. This cross was primarily to test the inheritance of rumplessness, and secondarily of black against red plumage color. THE RACES AS A WHOLE. Concerning the Rumpless Game see page 61. The Black Cochins are discussed at page 39. TABLE OF CHARACTERISTICS. No. Characteristic. Black Cochin Bantam. Discussed at page — Black-breasted Red Rumpless Game. Discussed at page — I General color. . . . Uropygium Iris color Vulture hock . . . Foot feathering. . Black 39 63 39 34 Red with some black. Absent 62 2 3 4 5 Present Dark brown . , Present Present 63 • • Red streaked with yellow Absent Absent MATERIAL. Mothers. — The Black Cochin Bantams Nos. 129 (fig. 26), 130, 131, and 132 were the same as those referred to at page 39. Father. — The Rumpless Game is No. 117, referred to at page 62 (fig. 46). 64 INHERITANCE IN POULTRY. RESULTvS. Only the first hybrid geueration has been produced. General Plumage Color. — Of 24 hybrids all were prevailingly black. Among 18 of those that hatched 8 showed some red. This red is chiefly found as a lacing on the hackle feathers or a peppering on the wing coverts,-*^ throat, t and outer margins of the remiges.]: This seems to point to the hypotheses that while black dominates over red the dominance is sometimes imperfect. When red occurs it occurs on those feathers that normally con- tain red in the Game, and on that part of the feather that is red in the Game. Uropvgium. — This is invariably present, apparently fully developed. Iris Color. — All the hybrids have dark-brown eyes; only one shows a trace of red. The hypothesis seems justified that in this case dark-brown iris pigment is dominant over red and yellow. Vulture Hock. — This is always absent. In only a single case§ are the feathers slightlj^ elongated on the hock. Foot Feathering. — Ever}^ chick that hatched has the foot and at least one toe booted. In some cases this booting is much reduced as compared with the Cochin parent. Booting is dominant, but not always completely' so. CONCLUSIONS. In this cross of black vs. red, black appears to be dominant, although im- perfectly so. The two colors do not blend, however, but red appears in a particulate fashion, usually in the parts of the plumage that have normally least black pigment. It is as if there were a struggle between the two pigments and red overcame black where black was weakest. The presence of tail in the first hybrid generation is confirmatory of the results of the preceding series. Rumplessness is apparentlj^ recessive. Brown iris color appears to dominate over the older red, and booting dominates over the ancestral clean-footed condition. * Nos. 589 (^ and 798 c^. t No. 587 ? • t No. 577 9 • § No. 651 J*. GENERAL DISCUSSION. 65 D. GENERAL DISCUSSION. INHERITANCE OF PARTICULAR CHARACTERISTICS. COMB FORM. The comb is a characteristic that has had its origin in the genus Galhcs. It consists of a mass of uncovered erectile tissue — a tissue present in many species of birds. The primitive form of the corab is the single comb seen in the wild species of the genus Gallus, and in most domestic races. This may be modified in two directions : First, in the direction of lateral repetition of the comb giving rise to the pea comb,* and, in an extreme case, to the rose comb (of which the walnut comb of the Malays is a special modifica- tion); second, in the direction of reduction of the modified comb producing the races with mere papillae (Houdan, Polish, I,a Fleche, etc.) or that are entirely combless (Breda fowl). That the rose comb is a modification of the same sort as the pea comb but carried to a greater extreme is indicated by the fact that the rose comb often shows five parallel ridges (instead of the more usual irregularly scattered papillae) and that in the female the ro.se comb sometimes consists of three ridges as in the male pea comb. When single comb (Minorca, fig. 4) and pea comb (Brahma, fig. 19) are crossed, pea comb is dominant (p. 35). The median ridge is, however, in the hybrid high for a pea comb and the lateral ridges are usually reduced (figs. 20, 21). When single comb (Leghorn) and rose comb (Minorca) are crossed, rose comb is dominant (p. 30). When .single comb (Minorca or Leghorn) is crossed with the paired rudiments of a comb found in the Polish and Houdan fowl, a Y-shaped comb results (pages 10, 22, 28, fig. 8). This Y comb is of great interest. It was obtained by Bateson and Punnett (1905, pp. 108, 112-114) in some of the offspring of (single-comb Leghorn x rose- comb Dorking), crossed with (single-comb Leghorn x walnut-comb Indian) ; and also in one of the offspring of a single-comb Leghorn crossed with [(single-comb Leghorn x walnut-comb Indian) x (single-comb Leghorn X rose-comb Dorking)]. In Bateson and Punnett's cases the splitting was evidently nearly complete, forming an 0-shaped comb, or the "cup comb" of Darwin (1876, Chapter VII). The Y comb was obtained also by Hurst (1905, pp. 133, 135, 138, 140, 146). This was a single split comb when Leghorn and Houdan were crossed, and a rose split comb when rose-comb Hamburgh and Houdan were mated. The interrelation of the different forms of comb — single, pea. walnut, Y, and V may, I think, be expressed in the following hypothesis : The pea comb and the walnut comb are composed of two element.s — a median single comb and a pair of lateral combs. This hypothesis is supported by the *The pea comb was doubtless a characteristic of the unknown feral ancestor of the Aseel-Indian group. But as the single comb is the dominant type in the known wild Jungle fowls the pea comb probably evolved from it. 5 66 INHERITANCE IN POULTRY. following evidence. First, of teratology. Extraneous paired papillae occa- sionally occur on the sides of the single comb in pure-bred races. These are known as ' ' side springs, ' ' and are considered by fanciers as grave ' ' defects. ' ' Now such side springs are morphologically equivalent to the lateral ridges of the pea comb. Second, there is the evidence of hybrid forms. Bateson and Punnett (1905 a) show that when pea comb and rose comb are crossed the second hybrid generation (FJ gives single comb, as well as pea and rose combs. This result may be interpreted as due to the fact that the gametes of a pea- combed bird have either a tendency toward side-springs (= pea comb) or they have no such tendency (:= single comb) ; and the gametes of a rose- comb bird have a tendency to produce two pairs of side combs (= rose comb) or else the}' have no such tendency (^= single comb). When two gametes without the side-comb tendency come together in F.^ a single comb is pro- duced. The necessity of assuming absence and presence of lateral combs strengthens the view that the pea comb is made up of two elements — median and lateral. If median comb and side-springs are distinct elements, then they should be independently inheritable. This result is realized on the one hand in the single comb, and, I think, on the other hand, in the cup comb (fig. 6), which consists of two side-springs without median comb. It is realized also in the V comb of the Polish fowl, which is a cup comb of which the anterior portion is typically not developed. That the V comb represents the posterior portion of a cup comb is supported by the fact that it is not uncommon to find not one pair of papillae merel5^ but two, three, or four pairs of papillae in Poli.sh fowl and in second-genera- tion hybrids. A row of three or four papillae on each side of the head is a close approach to a typical cup comb. The incompleteness of the cup comb where a V comb is produced may be due to various causes. In the Polish fowl the upturned nasal process and absence of a bony ridge over the nostrils appear to be the cause of the absence of a comb there, and we have seen (p. 17) that the only undissociable char- acteristics in the second-generation hybrids of Minorca and Polish are those of high nostril and rudimentary comb. The second cause restricting the development of the cup comb to its posterior limits is the presence of a median comb anteriorly ; this is the case of the ordinary Y comb. The Y comb is found in hybrids between .-^ ingle and V comb ; the anterior portion of the comb is not .suppressed here, because th^ bony roof of the culmen is com- pletely developed, and the very presence of a large median comb there prevents the development of the side-springs at the same niveau. In the development of the comb of the hybrid there is, as it were, a .struggle between the two elements of median and lateral combs. The Ycomb assumes a great variety of forms, running the entire gamut from a single comb on the one hand to (i) a cup comb or to (2) a pair of papillae on the other. I have already (p. 10) referred to the variation of the length of the stem of the Y, series INHERITANCE OF PARTICULAR CHARACTERISTICS. 67 (i),froiu 100 per ceuttouearlj-zero. The second hj'brids of Polish or Houdans crossed with single coiubs illustrate series (2). We begin with a single comb having its posterior one-sixth split ; next comes a comb having its posterior one-sixth split and anterior five-sixths single, but greatly reduced in height (fig. 50) ; next the same with the anterior portion reduced to an irregular carunculated mass having a slight median elevation (fig. 49), and finally a pair of papillae only (fig. 48). In this series we have a fading out of the median portion, pari passu with the enlargement of the nostril, but the persistence of the side combs unimpaired. The side combs have been unable to enter the territory from which the median comb has been driven, because that territory is likewise untenable for it. These two series sufficiently demonstrate that the V comb represents the posterior portion of the cup comb. That the cup comb represents merely the greatly enlarged lateral combs or side-springs is proven by the occasional presence of both median and cup comb on the .same individual. In some races, as in the English type of Hou- dans, the median comb typically appears lying between the pair of cup-like side-springs, resembling the trunk of a butterfly between its wings. Among the heterozygous combs of the second generation of Minorca x Houdan or Minorca x Polish hybrids, instructive examples of persistence of both single comb and side-springs are especially apt to occur. Figure 52 shows this condition ; there is a median comb anteriorly and a nearly typical pea comb posteriorly, except that the lateral ridges are atypically high. Thus the Y comb becomes explained as due to the presence of both single and lateral combs. The question now arises, Is it possible to explain on Mendeliau principles the production of a Y comb when median comb and lateral comb are crossed ? In accordance with such principles we should have to picture the gametes of the single-comb and V-comb parents as follows : Single comb. V-comb. Median element No lateral elements . . No median elements. Lateral elements. The allelomorphs are then median and no median, no lateral and lateral, and the positive characteristics are dominant. In the .second hybrid gen- eration the two dominant characters should be combined in nine-sixteenths of all cases ; the two recessive in one-sixteenth, and one dominant with one recessive in three-sixteenths + three-sixteenths of the cases. Another hypothesis is possible. Granting that the Y comb is no neomorph, but the sum of single and lateral comb, then the Y comb may be a case of particulate inheritance, the median comb being produced on the anterior and the lateral on the po.sterior part of the frontal region. In cases of particulate 68 INHERITANCE IN POULTRY. hybrids bred inter se, the offspring exhibits oue or the other of the parental conditions each in 25 per cent of the cases and the heterozygous condition in 50 per cent. To decide between these rival hypotheses we have to appeal to the statistics of occurrence of the different forms of comb. All cases (Series I and II} are combined in the following table, showing distribution in the second hybrid generation : Comb characteristic. Expected. Actual. On hypothesis of dominance. On hypothesis of particulate inheritance. Single comb Y comb . Lateral comb .... No comb Per cent. 18.75 56.00 Per cent. 25 5' 25 Per cent. 30.1 44-9 25.00 0? The foregoing table reveals several things. First, the actual distribution of comb form in the second generation accords better with the hypothesis of particulate inheritance than that of dominance of both single and lateral comb. That there is an excess of single comb and deficiency of Y comb is partly accounted for by occasionallj' counting a potentially Y comb but actually single (or nearl)- single) comb as a true .single. Secondlj^ the hypothesis of dominance demands the occurrence of a fourth form — pre- sumably no comb — in 6^ per cent of the cases. No combless fowl was raised to maturity, and the only possible cases were seen in still very j^oung or unhatched chicks. Probably no true combless bird appeared. From both of these considerations I conclude, provisionall}^ in favor of the theory that the Y comb is reproduced from the median and the lateral by particidai^ inheritance. NOSTRIIv FORM. The sum of results in Series I, II, and III (narrow x high nostril) gives : Generation. Narrow and intermediate. High. /. Actual. Expected. /• Actual. Expected. Fi F2 Fi X narrow 102 99 33 Per cent. 99.0 73-9 51-5 Per cnt. 100 75 50 I 35 31 Per cent. I.O 26.1 48.5 Per cent. 25 50 A close agreement exists between the percentage obtained in each genera- tion and the expectation on the Mendelian theory, assuming that narrow nostril is dominant. The statistics do not, how^ever, tell the whole story. In 36 per cent of the cases in the Fj generation the nostril was wider than in INHERITANCE OF PARTICULAR CHARACTERISTICS. 69 the ' ' narrow ' ' ancestor. Even in the F,^ generation nearly half of the ' ' nar- row and intermediate" were of the intermediate sort. This intermediate form is evidence that dominance is imperfect and segregation is incomplete. CEREIJRAI. HKRNIA. Cerebral hernia is, as already pointed ont, a typical monstrosity. The distribution of its occurrence in crossing is as follows : Crosses. Fi- F,. Fi X plain. Plain. Hernia. Plain. Hernia. Plain. Hernia. Minorca X Polish 66 24 16 106 *o 75 34 109 23 II 34 34 25 59 White L,eghorn X Houdan. . Houdan X Minorca . ... to Total Percentage 100 76.1 239 TOO * Excludiug one case of egg embryo with cerebral vesicle. t Excluding one egg embryo recorded as doubtful. Cerebral hernia is inherited in Mendelian fashion with plain head domi- nant. Nevertheless, many of the plain-headed hybrids have the frontal eminence abnormally high — -dominance is imperfect. CREST. The crest is independent of the cerebral hernia (pages 16-18). It is a widespread characteristic among birds, so common that it is not readily thought of as pathological but usually as ornamental. The distribution of iis occurrence in crossing is as follows : Crosses. F.. F^. Fi X plain. Plain. Crested. Plain. Crested. Plain. Crested, Minorca X Polish 70 II 6 • a 17 1 1 - 41 1 fi 6 White Leghorn X Houdan. . Houdan X Minorca . . Frizzle X Silky 01000 25 9 5 109 13 6 9 Total 54 12 15 Percentage 100 24 76 44.5 55-5 Crest is inherited in Mendelian proportions, and is dominant over crestless head. Even when the Silky is crossed with Gallus bankiva its crest is domi- nant (fig. 53). In this case the new characteristic, a positive variant, domi- nates over the ancient one ; but the crest is diminished in the first genera- tion ; dominance is imperfect. INHERITANCE IN POULTRY. WHISKERS OR MUFF. This is certainly a new character and a positive variant. The distribution of its occurrence in crossing is as follows : Crosses. F.- F,. Fi X plain. Absent. Present. Absent. Present. Absent. Present. Leghorn X Houdan Houdan X Minorca o o 24 II ? 26 26 5 5 II Total o 35 ? II Muffling is apparently dominant. BEARD. This is also a new, positive, variant. The distribution of its occurrence is as follows : Crosses. Fi- F.. Fi X plain. Absent. Present. Absent. Present. Absent. Present. Leghorn X Houdan Houdan X Minorca . ... 0010 23 10 33 ? 12 3 8 Total ? 12 3 8 Beard is apparently dominant, but often imperfectly so. FEATHER FORM. Silkiness is a new characteristic and, approximating as it does the juvenile down condition, a negative one. When a Silky is crossed with a Jungle fowl the offspring are plain. Silkiness is recessive to non-silkiuess — the retrograde to the progressive type. Frizzling is likewise a new characteristic — a positive character added to the perfect feather. The distribution of the occurrence of silkiness and friz- zling is as follows : The Frizzle fowl used were doubtless heterozygous. When non-frizzled birds are crossed inter sc they produce only plain offspring. Frizzling is dominant over non-frizzling — the progressive over the primitive. INHERITANCE OF PARTICUI.AR CHARACTERISTICS. 71 UROPYGIUM. Rumplessness is a new characteristic and a typical negative variant. The distribution of its occurrence is as follows : Fi- Non-rumpless. Rumpless. Leghorn X Rumpless Game Cochin X Ruuipless Game 23 19 7 3 *o Frizzle X Rumpless Game Nankin X Rumpless Game Total . Percentage 52 IOC * One egg embryo doubtfully rumpless. The new, negative characteristic is here completely recessive. TAIL-I,ENGTH. The long tail of the male Tosa fowl is a new, positive variant. The dis- tribution of its occurrence in male hybrids is as follows : Crosses. Fi- F,. Short. Long. Short. Long. Tosa X Cochin Brahma X Tosa Total 3 *i6 ? ? 19 * Tlie tails are, perhaps, more properly intermediate. While still growing at date of record, they grow slowly. The new, positive characteristic is doubtfully dominant, possibly inter- mediate {cf. fig. 34). VUI^TURE HOCK. This bundle of strong feathers constitutes a new, positive characteristic. The distribution of its occurrence is as follows : Cros.ses. F,. F,. Absent. Present. Absent. Present. Minorca X Brahma 7 12 12 9 t3 All j III *o (Small) I t3 trades. 16 t 20 Leghoru X Brahma Black Cochin X Leghorn Leghorn X Buff Cochin Tosa X White Cochin Brahma X Tosa Black Cochin X Rumpless Game. * One shows trace of enlargement ot teathers. t Females with vulture hock ; males without it. X Seven recorded as slight. S Oue case of trace of elongation of feathers. 72 INHERITANCE IN POULTRY. The result is peculiar. Usually the vulture hock is absent iu the first hybrids, indicating its recessiveuess. In crosses with a particular race — Tosa fowl — however, there is no recessiveuess. It is probable that the Tosa fowl is heterozygous iu respect of this characteristic. The new character- istic is recessive, but imperfectly so. FOOT FEATHERING. Foot feathering, as the discussion on page 34 indicated, is a positive vari- ation, new to Gallus, but not of a pathological sort. Common among wild, scratching birds, its occurrence in Gallus may be regarded as a case of de- gressive variation (de Vries). The distribution of its occurrence is as follows : Crosses. Fi- F.. Non- booted. Booted. Non- booted. Booted. Minorca X Dark Brahma Leghorn female X Dark Brahma male Dark Brahma female X LfCg- horn male Black Cochin X Leghorn Leghorn X Buflf Cochin Tosa X White Cochin Dark Brahma X Tosa Frizzle X Silky Black Cochin X Rumpless Game . Total I 4 3 3 40 15 25 20 26 7 22 15 21 7 48 II 191 7 48 The foregoing statistics tell only a part of the story. Booting, when present, is frequently much reduced ; one may regard absence of booting as the extreme condition. Booting is dominant, but usually imperfectly so. EXTRA TOES. The extra toe is a positive variation of a teratological sort, tion of its occurrence is as follows : The distribu- Pi. F,. Fi X normal. Crosses. No extra toe. Extra toe. No extra toe. Extra toe. No extra toe. Extra toe. Houdan X Leghorn Houdan X Minorca Frizzle X Silky 6 6 7 19 31 15 23 17 17 6 ■■ 6 17 8 Total 69 17 S Percentage 21.6 78.4 73-9 26.1 68.0 32.0 INHERITANCE OF PARTICULAR CHARACTERISTICS. 73 These results are peculiar. If both normal-toed and extra-toed ancestors were heterozygous in respect to toes, we should expect the result obtained iu Fp It is quite possible, though not probable, that this is true. Then extra toe would be dominant, although sometimes so imperfectlj^ so as not to appear. In F, the parents were normal-toed, either because "normal" is recessive or because it is imperfectly dominant. All offspring should be normal-toed in the one case or give 100 per cent to 75 per cent extra-toed in the other. The result is not in accordance with either hypothesis. If there is any dominance in this generation it is of the 7iormal toe. Bateson and Saunders (1902, p. 124), while concluding that extra toe is dominant, find "that the recessive foot character may sometimes dominate." Hurst (1905, p. 150) also got, in a cross between Iveghorn and Houdan, some normal-toed ofifspriug which, interbred, produced extra-toed progeny. He concludes that a usually dominant character may recede in certain individ- uals. There is danger here of straining Mendel's law. It is better to hold ' ' explanations ' ' in abeyance until the matter of inheritance of polydactylism has been more thoroughly investigated. Certainly the facts of inheritance of polydactylism in man can hardly be explained on Mendelian principles (Davenport, 1904). Polydactylism is at least not recessive. The new, posi- tive, pathological characteristic holds its own against the older one. SKIN COLOR. The epidermis of poultry is everywhere covered by feathers except on the beak, face, and feet. The naked portions may, however, have a different color from the covered ones ; consequently the correlation between general skin, beak, and foot color, although not absent, is not close. Thus, although the yellow beak and foot of the lyCghorn are correlated with its yellow skin, the black legs and beak of the Black Minorca are not accompanied by a black skin. Not all exposed parts, even of the skin, are of one color, for the face, at least, maybe red or white when the legs are black. Color of beak and foot are, on the other hand, closely correlated, individual variations of the one being usually associated with corresponding variations of the other. This correlation is doubtless the result of the similar cornification of the skin of beak and foot, whereas (excepting races with opaque white face) the vascular face and earlobes are white or red, according to a less or greater blood supply in them. The pigmentation of the epidermis of poultry falls into three classes : (a) Without pigment or white ; [U) yellow ; {c) black. White skin is the com- monest, even among poultry with black plumage and feet. Yellow skin is found in the Asiatics, derived from the Aseel- Malay ancestry, and is a char- acteristic of the White Leghorn. Black pigment occurs in the skin of the Silky fowl and the Negro fowl. Black pigment is to be regarded as a new variant and of the nature of a pathological sport — melanism. When black 74 INHERITANCE IN POULTRY. skill is crossed with white, black — the new, positive, pathological character- istic — is dominant (page 60). MANDIBLK COLOR. The prevailing types are black, willow, yellow, and white. Black is the primitive color on the Jungle side ; yellow, on the Aseel group. When horn (Houdan) and yellow (Leghorn) mandible colors are crossed, the first genera- tion shows the yellow of the Leghorn, which is dominant. When, however, the black beak of the Minorca was crossed with the yellow beak of the Brahma, the dark color dominated. The potency in the hybrid of beak color seems to follow this series : Black, yellow, horn. The most positive char- acter, black, dominates all. FOOT COLOR. Four principal types are to be distinguished — willow, black, yellow, and white. Willow is primitive and white the most aberrant. The results are based on still insufficient data, but so far as they go they indicate that willow is dominated by yellow (p. 54), yellow by white (p. 24), and white by black (p. 28). The newer, negative characteristic, white, is dominant over the older yellow, but the new, positive characteristic of melanism dominates all. IRIS COLOR. Of the various forms, pearl to yellow is characteristic of the Aseel type ; red, of the Game or Gallics bankiva type. Black has become associated with black plumage. The results, subject to revision, indicate that in poultry, as in man, iris color rarely blends, that red dominates pearl (page 38), and that dark brown dominates red. The new, positive variation of melanism seems to dominate all, although not alwaj's perfectly. EARLOBE COLOR. Red is primitive in both groups. White is a new variation, which is prob- ably due to fat or other particles in the skin, and is consequently positive. Only in extreme cases is red wholly eliminated from the earlobe. In three series of crosses (V, VI, and X) of the red-lobed Dark Brahma and a white (and red) lobed race the earlobes were prevailingly red, but had some white at their centers. Likewise, in two series of crosses (VII and VIII) of the red-lobed Cochin and a white- lobed Leghorn, red dominated in the hybrids, but did not always perfectly exclude white. Red is apparently dominant, but very imperfectly so ; some cases rather indicate particulate inheritance. GENERAL PLUMAGE COLOR. The original plumage as exhibited in the Jungle fowl is largely black and red ; that of the Aseel type sometimes contains much white ; but the pure white plumage must be regarded as a new negative variant. The outcome of crossing is complex. INHERITANCE OF PARTICULAR CHARACTERISTICS. 75 White vs. Dark — Three different results may be, under differing condi- tions, obtained. Dominance of White. — This is the usual result. Tu'o White Leghorns crossed by a black Minorca produced only white hj'brids, but the female hybrids, at least, had some black feathers. White Leghorns crossed with Houdans gave only v/hite. White Leghorns crossed with a Red-backed Game had white offspring with some buff on breast. On the other hand, the white color of the Silkj- dominates over the dark color of the Frizzle CSeries XI) in about only 23 per cent of the hybrids. Bateson and Saunders (1902, pp. 108-109), dividing all hybrids between black and white parents into those of light type and those of dark type, conclude that the former are to the later as 3.1 to i. Bateson and Punnett (1905, p. 117) conclude that offspring of a pure white parent with colored or heterozygous (mixed) birds are practicallj^ always prevailingly white. Hurst (1905, pp. 146-149) gets chiefly white birds from crosses of White Leghorn hens with black or mottled males. The exceptions may be due to the impurity of one of the females. Barring. — No barring resulted from crossing White Leghorn withHoudan or black Minorca, or vSilky with Frizzle. On the other hand, all males, and only males, were barred in the hybrids of Tosa x White Cochin, and in the hybrids of White Leghorn Bantam and Rumpless Game barring occurred, but among males only. Of 26 hybrids between Black Cochin and White Leghorn, 8 were barred black and white, and these belonged equally to the two sexes. Of 11 dark h5'brids obtained by Hurst (1905, p. 133) frooi White Leghorn x Houdan, 6 developed into black females and 5 into cuckoo males. Apparentl}^ barring ("cuckoo marking" of the English) is asso- ciated with maleness. This result is curious enough, for, as Darwin pointed out, in the ancestors of domestic poultry barring (or rather penciling) is confined to the female sex. Barring is a heterozygous condition found in hybrids from a white and a black parent. It is provisionally regarded as a form of particulate inherit- ance as opposed to the alternative inheritance of the Leghorn x Minorca cross. This heterozygous condition when interbred usuall}- breaks up into white, uniformly pigmented, and barred again, as in the case of the Tosa X White Cochin hybrids (p. 49). This form has in certain cases, as in the Cuckoo Dorkings and in the Dominiques — ancestors to the Plymouth Rocks — become truly mosaic, transmitting the mixture of qualities pure. The method of fixing a heterozygous quality is still unknown to science.* *The experience of breeders of mice and guinea-pigs shows that white may be due to the absence of an oxidizing ferment necessary to the bringing out of the color potential in a chroaiogenic substance (<■/. von Fiirth, 1903). If the chrouiogen is present the addition (by crossing with a pigmented individual) of the ferment will reveal in the hybrid ofY- spring the colors and pattern latent in the white parent. Working on this hypothefis, we can judge of the latent patterns in the White Leghorn bantams and draw conclusions 76 INHERITANCE IN POULTRY. Andalusian Coloration. — Among the ofiFspring of a White Leghorn and a Black Minorca two adult blue fowls were reared (fig. 54) . The coloration was that of the Andalusian "breed." It consisted of a minute patchwork of black and white pigment. Such a blue coloration is common in barn-yard fowls. It results, according to the testimony of breeders,! from crossing black and white. The special conditions which determine whether the offspring of a white and a black parent shall be all white or barred or blue have not yet been determined. The solution of this problem oflfers one of the most interesting fields for future investigation (p. 30). White vs. Buff. — Both colors are novel ; the former is probably a nega- tive mutation ; the latter has been extracted from the original game colora- tion of fowls. The hybrids are prevailingly white, and white may be regarded as dominant. Nevertheless, this dominance is imperfect, for in half of the offspring buff is more or less evident. It is found diffused over the back, wings, and breast as in " pile" Games. On the whole, white is less strongly dominant over buff than it is over black (Hurst, 1905, p. 134). Black vs. Red. — The red coloration is ancestral ; the solid black is novel and positive — a melanic condition. The hybrids between Black Cochin and Red-breasted Game are prevailingly black, but about half of them show red lacing on the hackle feathers or a red peppering in those places where red is displayed by the Game. Black is dominant over red, but imperfectly so. COLOR OF TOP OF HEAD. In the white-crested Black Polish the feathers of the top of the head are in striking contrast to those over the rest of the body. That the crest is not necessarily white is proven by the existence of a black-crested race. Hybrids between the Minorca, whose head is wholly black, and the Polish give (p. 15) chiefly black feathers in the males, the females, however, still showing as to what pigmeated ancestors they may have had. They were used iu five crosses, as follows : (i) Black Cochin X White Leghorn; (2) White Leghorn X BnflF Cochin; (3) White Leghorn X Black-breasted Red-backed Game ; (4) White Leghorn X Dark Brahma ; (5) Dark Brahma X White Leghorn. Taking all offspring together, about 50 per cent (48.5) are white or nearly so. All crosses exhibit barring, together in about one-quarter (26,5) of the cases, and also black and buff or red. It seems probable that all of these pigments and the barred pattern are latent in my White Leghorn bantams. These conclusions are supported by breed- ing the White Leghorns inter se, when, in addition to white offspring, a black and a barred were obtained (p. 40). Similarly among the second hybrids between theTosa and White Cochin Bantam there appeared a male and a female resembling in plumage colora- tion the Partridge Cochins (p. 49). This coloration probably lay latent in the gametes of the White Cochin. t Compare Darwin (1876, I, Chapter VII ; 1S94, I, p. 270) ; Wright (1902, pp. 291, 292, 317. 301. 399. 401, etc.) ; Bateson and Saunders (1902, p. 131) ; Bateson and Punnett (1905, p. 126). When blues are interbred, the offspring are either white or black or blue. Even in the Andalusian " breed " the blue coloration has never become fixed. INHERITANCE OF PARTICULAR CHARACTERISTICS. 77 white in their crests. The hybrids crossed back on the Minorca give nearly 100 per cent black heads. Black is dominant, but imperfectly so ; the nega- tive characteristic is recessive. The dominant character is less perfectly dominant in the female sex than in the male. COr.OR OF HACKI.es— HACKLE LACING. The color of the hackle feathers and the correlated saddle feathers in birds of broken color usually differs from that of the rest of the plumage. This peculiarity of the hackle coloration is an old character, since it is exhibited by the Jungle fowl, and was probably in the ancestor of the Aseel-Malay group. The feathers are laced with a lighter color than the center. In crosses between Minorca and Dark Brahma, and White Leghorn and Dark Brahma the solid color (black or white), the new, positive character- istic, dominates over the lacing. Nevertheless, in the Minorca X Dark Brahma hybrids the feathers of the nape are frequently faintly laced with gray. The black is imperfect!}' dominant. WING COLOR— RED WING COVERTS. The male Jungle fowl has red on the upper wing coverts, and doubtless the male of the ancestors of the Aseel-Malay group had also. The male hybrids between the Dark Brahma and the Black Minorca on the one-hand and the White Leghorn on the other usually shovv' red on the wing coverts, although there is no other red in the plumage. Red on the wing coverts is probably dominant, but it is much reduced. TAIL COLOR. Although the tail feathers are derived from a distinct feather tract, and in broken-colored birds are usually without the red of the wing, yet tail color does not seem to be a unit character ; in inheritance it follows the rest of the body plumage. On the other hand, in breeding buff varieties black persists in the tail feathers longer than in the others. This case resembles the per- sistence of black at the extremities of the legs of white or red rabbits (Castle, 1905). .SHAFTING. The female Jungle fowl has a light shaft to the feather. The same is true of the Tosa fowl and some Games. Light shafting is a primitive characteristic of the female. In the female hybrids between the Tosa fowl and White Cochin the shaft- ing is greatly broadened, and this is the principal modification of the plumage color. In female hybrids of the Tosa fowl and Dark Brahmas the shafting of the feathers of the back and wing coverts is striking, and some shafting appears in two of the males, probably transferred from the female (p. 54). Apparently shafting is dominant. 78 INHERITANCE IN POULTRV. BODY LACING. This character is uot fouud in the Jungle fowl, but may have been derived from the penciling of the Aseel-Malay group. In male hybrids between the Tosa fowl and the Dark Brahma it occurs, derived from the latter (p. 54). It appears to be dominant. PENCILING. This is an ancient feminine characteristic, best marked in the Aseel- Indian group (p. 53). It is found particularly well developed in the Dark Brahma female. In the female hybrids between that race and the Tosa fowl penciling is well developed ; it is dominant. GENERAL TOPICS IN INHERITANCE. UNIT CH.\RACTERS. Taxouomic descriptions of plants and animals give a list of their specific characteristics (Merkmale, caracteresu These comprise for the most part only tha external characteristics, but a similar list might be made for internal characteristics. In addition to specific characteristics, those of a higher order (such as generic, etc. ) and those of a lower order (such as varietal) may be enumerated. Such characteristics are, in first approximation, unit char- acters. They are of prime importance, because the whole problem of evolution is that of the origin and significance of the various unit characters of the body. The theory of the unit character is associated with that of its bearer in inheritance. Darwin (1876) and later de Vries (1889) designated as such bearers particles of the nuclear material named " pan genes." "Changed numerical relation of pangenes is the basis of fluctiiating variability ; dis- placement (Umlagerung) of pangenes in the nucleus conditions retrogressive and degressive mutations ; while the formation of new kinds of pangenes is necessary to the explanation of progressive mutations " (/. c, those exhibit- ing altogether new characteristics). The two main hypotheses of the origin of unit character ire that of de Vries and that of Weismann. De Vries sets forth his hypothesis at the very begin- ning of his great work, " Die Mutatioustheorie." His words may be thus translated : As mutation theory I designate the doctrine that the characteristics of organisms are built up of units that are sharply separable one from another. These units can be united into groups, and in related species the same units and groups recur. Transitions, such as the external forms of plants and animals exhibit in such numbers, exist between the units as little as between the molecules of chemistr}- In the realm of the doctrine of descent this principle leads to the conviction that species have proceeded from one another not continuously but by steps [nicht fliessend, aber stufenweise]. Each new unit added to the older ones constitutes a step and separates the new form, as an independent species, sharply and fully from the species whence it arose Die neue Art ist somit mit einem Male da ; sic entsteht aus der friiheren ohne sichtbare Vorbereituug, ohne Ubergange. GENERAL TOPICS IN INHERITANCE. 79 Weismana, on the other hand, is only less clear in expressing his hypoth- esis. He accepts, of course, the idea of unit characters, each of which is rep- resented in the germ cells by a "determinant." "We called," he says (1904, I. P- 369), " determinants those parts of the cjerm- substance which determine an ' hereditary character ' of the body ; that is, whose presence in the germ determines that a particular part of the body, whether it consists of a group of cells, a single cell, or a part of a cell, shall develop in a specific manner, and whose variations cause the variations of these particular parts alone. ' ' The " hereditary parts " may be small or " large regions, whole cell masses of the body, which in all probability vary only en bloc, as, for instance, the milliards of blood cells in man, the hundreds of thousands or millions of cells in the liver and other glandular organs, the thousands of fibers in a muscle, or of the sinews or fascia, the cells of a cartilage or a bone, and so on. In all these cases a single determinant, or at least a few in the germ plasm, may be enough." For Weismann (1904, II, p. 151) the ultimate source of all hereditary variations is the variation of the representatives of the unit char- acters in the germ plasm. " If I mistake not," he says, " we may at least say so much, that all variations are, in ultimate instance, quantitative and that they depend on the increase or decrease of the vital particles, or their constituents, the molecules What appears to us a qualitative varia- tion is, in reality, nothing more than a greater or less different mingling of the constituents which make up the higher unit ; an unequal increase or decrease of these constituents, the lower units." The cell changes its consti- tution when the proportion of its component parts " is disturbed, when, for instance, the red pigment granules which were formerly present, but scarcely visible, increase so that the cell looks red. If there had previously been no red granules present, they might have arisen through the breaking up of certain other particles — of protoplasm, for instance, in the course of metab- olism —so that, among other substances, red granules of uric acid or some other red stuff were produced. In this case, also, the qualitative change would depend on an increase or decrease of certain simpler molecules and atoms constituting the protoplasm-molecule." In criticism of the foregoing it may be said that a variation in the number of atoms in a protoplasmic molecule is certainly also a qualitative change — a mutation. The only real dift'erence between Weismann and de ^'■ries depends on the extent of the mutative modification, whether progressive or complete from the beginning ; but this is a real difference, for the latter view is required by the theory of immutable unit characters. The former view is not in harmony with such a theory. Conversely, if it appears that there are immutable unit characters, then the theory of evolution by saltation is necessary ; if unit characters are modifiable, then species may have arisen gradually. 8o INHERITANCE IN POULTRY. The result of the breeding experiments described herein bears upon this discussion. No other group, I imagine, exhibits so many characteristics as poultr}' ; of the comb alone there are half a dozen forms. The forms of feathers and their color patterns are numerous. These forms are sharply marked off from one another for the most part ; moreover, when two char- acteristics are crossed the result is rarely a blend. This was a great surprise to me, as I had anticipated that blends would be the rule ; and, overwhelmed by the facts, I embraced at once the theory- of immutable characteristics. That there are unit characters in poultry can not be doubted. When single and V comb are crossed and progeny obtained all with a Y comb, how con- vincingly do the second hybrids reproduce the single comb in some individuals and the V comb in others ! Though the cerebral hernia and its associated great crest may disappear in the first generation of hybrids, how beautifully do they reappear in one-fourth of the offspring of such hybrids ! How in- structive is it to see perfectly plain feathered offspring arising from a frizzled pair, or in a Black Minorca X Dark Brahma white-laced hackles appearing in an otherwise dead-black plumage ! Truly we may hope, as in chemistry, to make various kinds of molecules by the proper admixture of our atoms — the characteristics. Even in man such non-blending characteristics are evi- dent. One of the most famous is the Hapsburg lip or chin, which from the fifteenth century has persisted to the present day despite infusion of new blood during fifteen generations.* Another striking case is that of hypo- phalangia in man, described by Farabee (1905). In the four or five gener- ations studied, there has, he states, "never been a single instance of partial inheritance, but in all cases all extremities have been aiTected in precisely the same way." While admitting, thus, the reality of unit characters, the further study of the evidence of hybridization in poultry has led me away from the conception that they are rigid and immutable as atoms are, which may be combined and recombined in various way and ahvays come out of the process in their pristine purity. This is by no means the case. Very frequently, if not always, the character that has been once crossed has been aflfected by its opposite with which it was mated and whose place it has taken in the hybrid. It may be extracted therefrom to use in a new combination, but it will be found to be altered. This we have seen to be true for almost every char- acteristic sufficiently studied — for the com^b form, the nostril form, cerebral hernia, crest, muff, tail length, vulture hock, foot- feathering, foot color, ear- lobe, and both general and special plumage color. Everywhere unit char- acters are changed by hybridizing. How does this fact bear on the rival theories of evolution ? It has an im- portant bearing on them. It is not in accord with the statements of de Vries Cf. V . A. Woods, 1902-03. GENERAL TOPICS IN INHERITANCE. 8 1 quoted above : " The characteristics of organisms are built up of units that are sharply separable one from another," and "Transitions exist between the unit as little as between the molecules." Single comb is one unit and pea comb is a different unit, but they are not sharply separable. Crest and no crest are units, but they run into each other in hybridizing. Unit char- acters may show transitions, and, if so, they viay have originated gradually, so far as I see. It does not follow that they must have originated gradually. ai.te;rnative, particulate (mosaic), and blending inheritance. Doubtless Darwin's statement that crossed characters usually blend is still the prevalent view. Much if not most biometric work in heredity has been made on this basal assumption. I may say that I began my experi- ments prejudiced in favor of this view\ The results that have been recorded in the foregoing pages indicate that probably in general typical blending of characters is rare. Excepting char- acters like general form of the body, which are doubtless not units, but complex. I have, indeed, seen no single case of a typical blend. A fusion of characters is a rather rare phenomenon. Human skin color is the one striking case. One can but wish we had more careful data on inheritance of human skin color in successive generations. Other human characteristics show alternative inheritance. This is strikingly true in Farabee's family of hypodactyls cited above. It is said to be true of eye color and probably of the states of general pigmentation known as blonde and brunette. The following characters of poultry show alternative inheritance : Comb form. Uropygium. Earlobe color. Nostril form. Tail length. General plumage color Cerebral hernia. Vulture hock. (sometimes). Crest. Booting. Color of hackles. Muff. Extra toe. Wing bar. Beard. Color of mandible Shafting. Frizzling. and foot. Body lacing. Silkiness. Iris color. Penciling. The following characteristics show particulate inheritance : Iris color (sometimes?). White and black, producing barring (Series VI, VII, IX, XII). White and black, producing blue (fig. 54) (a fine mosaic of white and black). It is too early yet to interpret the cases of particulate inheritance. It is a striking fact that, excepting the Tosa X White Cochin cross, all my barred birds reared to maturity had the White Leghorn Bantams as mother or father. Now, as repeatedly observed, these bantams were heterogametous. It is possible that they contain barred blood in the "fixed" condition. Aside from the fact that they throw a certain proportion of barred birds, this conclusion gains support from the fact that the wing coverts of the male are 6 82 INHERITANCE IN POULTRY. obscurely barred with dusty bands, althougli, on the other hand, this barring may be merely the badge of hetero/.ygotism. It is possible, therefore, that the barring in the plumage of the White Leghorn Bantams is transmitted as an alternative characteristic. The case of the barred descendants of the Tosa X White Cochin is more difficult. I am not yet prepared to go so far as Correns (1905"*, p. 13, note) when he says: "Wo Mosaikbildung als Regel bei einem Bastard auftritt, war sie scbon in einem der Eltern oder in beiden, aktiv oder latent, vorhanden," Naturally, attention was directed chiefly toward evident qualitatively marked characteristics. Such do not blend. The fact that for the most part a characteristic does not blend when crossed with its allelomorph is of the highest importance for the theory of evolution. If blending were uni- versal a new characteristic must inevitably become quickly swamped by intercrossing with the parental characteristic. Since the new quality does not blend, it need not be swamped, even when there is no special isolation. INHERITANCE OF SPECIFIC VS. VARIETAL CHAR.\CTERISTICS. A distinction between specific and varietal characteristics is made by Nageli (1884, p. 247) and by de Vries (1902, p. 141 ; 1905, p. 141). Fol- lowing de Vries, a specific characteristic is a wholly novel one acquired by the race — one which stamps its possessor as an elementary species, A varietal characteristic is sometimes positive (/. ' of the fowl crossed ; second, general distribution of barring among the offspring ; third, proportion of different forms of plumage pattern in generations beyond the first. The cross between Tosa fowl and White Cochin gave barred birds. If the barring were latent it must have lain in the Cochins — the form without visible pattern. It is fairly certain that neither of the ancestors of domestic fowl was barred ; hence if the barring determinant existed in the Cochin bantam it must have been introduced by a recent cross. Bantamizing of Cochins is effected by crossing with some bantam race, but until recently no barred bantams have been created. It is therefore highly improbable that a barred bird was used to bantamize the Cochins. While it is possible, it is improbable that the White Cochin contained a barred determinant. Second, barred races have the two sexes equally barred, but our hybrids are barred in the male only ; consequently barring here acts like a neomorph. Third, on the theory of atavism we should expect to get in the second hybrid generation : Coloration of second generation. Atavism Particulate theory. theorj'. Actual. White Pigmented and barred Per cent. 25.00 56 25 18.75 Per cent. 25 50 25 Per cent. 28 48 24 Pigmented and not barred The actual proportions of the three types accord much better with the particulate inheritance theory than with that of atavism, but the total number of offspring is insufficient to give certainty. It may be concluded that while the evidence does not exclude the atavism theory of the cropping out of barring, it favors the theory of particulate inheritance. The case of the hybrid between single and V comb rests on more extensive data. These are set forth on pages 10-12, and are less favorable to the ata- vistic theory than to the particulate theory. The other heterozygous forms have been less carefully studied. They are the blue, Andalusian (fig. 54, pi. xvii), plumage color resulting from a white and a black crossed, and the case of the down of the hybrid Minorca x Dark Brahma chicks. This is black like the Minorca, but lacks the white of the chicks both of that race and of the Dark Brahma. The Andalusian 90 INHERITANCE IN POULTRY. hreed has been discussed by Bateson and Punnett (1905, p. 126), and they find, what is the universal testimony of breeders, that (as stated also at page 76) the blues bred inter se produce some white and some blacks, but still more blues. Until more complete statistics have been gained on the pro- portions of colors in the offspring, the interpretation of blue must remain uncertain. Hybrid forms are, then, frequently cases of particulate inheritance in which the hybrid gametes are not mosaic ; consequently whenever ' ' pure ' ' offspring are produced, as in Fj, these reassume the character of the pure race. In some cases, as in the cuckoo Dorking and the Dominique (from which our barred Plymouth Rock has been derived), the heterozygous form of barred plumage has become fixed, so that only barred offspring are pro- duced. A mosaic gamete has been created. The blue coloration has never yet been fixed as a permanent hybrid form. The method of fixing a hj-brid form is urgently in need of investigation. REVERSION. This term has been used rather loosely in the past for the appearance in hybrids of characteristics not visible in the immediate parents of the hybrids and often belonging to remote ancestors. Darwin (1876) made much use of this term in describing his results. He believed that the occurrence of " reversion " gave a useful key to ancestry. It is worth while to consider his observations and experiments. He mentions the fact that " purely bred Game, Malay, Cochin, Dorking, Bantam, and Silk fowls may fre- quently or occasionally be met with, which are almost identical in plumage with the wild G. baiikiva.'' But does this indicate anything else than that this type of coloration has persisted in certain primitive races, like the Game, and has been transplanted from them to the new races ? Darwin crossed a black Spanish cock with various white and white- and-black hens of pure breed. The offspring of this cock crossed with a silver-spangled Polish hen and with a white Cochin hen showed no sign of reversion to the red color of G. ba?ikiva. The male offspring of a spangled or silver Hamburgh hen showed white in the hackles and a reddish yellow on the saddle. Darwin regarded this as a " first sympton of reversion ; " but in the first of these peculiarities the hybrid resembles G. bankiva less than the Dark Brahma. The offspring of a white Game hen with the Spanish cock was at first snow white, but eventually produced the "pile" coloration. Darwin regards this as a partial reversion to G. bankiva ; but it is equally possible that the reversion is only to a pile coloration that is latent in the white from an earlier cross and is brought out when the white is crossed with a dark color. But Darwin's most remarkable hybrid was the offspring of a white Silky hen. Of two cockerels one was black (with light laced hackles) ; the other resem- bled closely a Jungle cock. Darwin admits that the case is extraordinary, GENERAI, TOPICS IN INHERITANCE. 91 but it was duplicated by Mr. Tegetmeier. This experiment certainly should be repeated, and I have arranged to repeat it next season. One of the best cases of reversion is the gray coat of a hybrid between a white and a black mouse. We now know, however, that even a "pure race ' ' of white mice may carry gray as a latent characteristic that first be- comes patent on crossing. In view of such facts cases of " reversion " to a remote ancestor must be critically examined. If the " reversion " be not a neomorph, it must have been handed down without break in the germ plasm from an ancestor possessing the characteristic. PURITY OF GAMETKS. The dogma of purity of the gametes, the second corner-stone of Mendelism, asserts that while the unripe germ cells of a hybrid having antagonistic or alternative characteristics A and A' contain representatives of both A and A' , yet the ripe germ cells of such a hybrid contain representatives of either A or A' , and not of both. Thus the ripe germ cells (gametes) are pure in respect to a given characteristic. They gain this purity, it is supposed, dur- ing the maturation period, the period when the reduction division of the chromosomes occurs, and when in each cell division one-half of each chro- mosome moves bodily to one daughter cell and one half to the other. The theory assumes, of course, that characteristics A and A' , being derived from different parents, inhere in different chromo^:omes. Let us assume that our hybrid has eight chromosomes, four derived from each parent, thus : .0 • o o • in which the black dots represent chromosomes of maternal origin ; the circles chromosomes of paternal origin. If all maternal chromosomes contain the determinant a then purit}^ of the gametes demands that all such go to one gamete and all of the chromosomes of paternal origin go to the other, and that such is their behavior has in fact been assumed by Cannon (1902). But that would result in the extracted pure individuals of the second hybrid generation being like their grandmother or their grandfather in all charac- teristics, which is not the case. If we assume that some only of the maternal chromosomes, such as are represented by the small dots, contain the deter- minant a, then these may be a.ssociated with an}' of the paternal chromo- somes excepting those that contain the determinant a' . Such a selection of chromosomes so as to exclude from the ripe gamete chromosomes containing both the alternative characteristics is quite possible, owing to the fact of synapsis, in which the homologous chromosomes from the two parents unite in pairs, as shown in the figure, in such a way that both can not pass to the same gamete. The foregoing hypothesis of Sutton (1902, 1903) and Boveri (1902) would account for perfect purity of gametes. But it is clear that gametes are not 92 INHERITANCE IN POULTRY. wholly pure, since the characteristics in second generation hybrids are rarely exactly like those of their grandparents ; consequently various additional hypotheses have been offered accounting for this feature. Hacker (1904) points out that chromosomes do not pass from cell to cell unchanged except for their growth and division. During the resting stage "of the nucleus it disappears. The new nucleus which arises in the position of the old is at first small ; it arises inside of the old chromosome as a spore arises in the mother cell ; its material has been derived from a part only of that of the mother chromosome ; the remainder goes to form part of the cytoplasm. Though chromosomes from different parents tend to separate to distinct gametes, still all gametes are infected by each kind of characteristic." McClung (1905, p. 329) assumes, more vaguely, a mutual influence of synap- tically paired chromosomes in the prophase of the first spermatocyte. A different suggestion is offered by Ziegler (igos). He assumes that each chromosome of maternal or of paternal origin carries determinants of all characteristics. After maturation all gametes contain the same number of chromosomes, but the proportion in them of chromosomes of paternal and of maternal origin varies. Gametes rarely contain exclusively maternal or paternal chromosomes, but whenever the proportion from one parent is high the gamete acts as though it contains exclusively the gametes of that one ancestor. If two gametes that are prevailingly paternal unite in a zygote the resulting hybrids (of the second generation) show all the grand-paternal characteristics. The difficulty with this hypothesis is that, like Cannon's, it does not account (any better than the first hypothesis) for the diverse combinations of characteristics shown in the second hybrid generation. Still another suggestion has been made by Morgan (1905). It is that the gametes are not pure, but contain determinants of both allelomorphs a and a', and that one of these dominates in half of the gametes and the other in the remaining half. The advantage of this hypothesis is that it accounts for latent dominant characters in recessive individuals. This hypothesis assumes that the gametes of hybrids are alwaj^s impure, and that this im- purity can not be got rid of. This seems to me to be contrary to experience. Moreover, except for the explanation that it offers of latency — which has been accounted for on other grounds by Cuenot — it offers no practical advan- tage over the theory of pure gametes. From the foregoing diversity of hypotheses it is evndent that we lack a fully satisfactory cytological explanation of the facts other than that of purity — the fact of imperfect dominance and the fact of particulate inheritance com- bined with purity in the second hybrid generation. Perhaps it will suffice to suppose a restricted purity of gametes such that the determinant of a characteristic may become infected to a slight degree by the presence of its allelomorph. GENERAL TOPICS IN INHERITANCE. 93 COMPARISON OF RECIPROCAL CROSSES. There is a notion among breeders of poultry that the father and the mother contribute different qualities to the offspring ; and if the cytoplasm carries any hereditary tendencies this result is to be expected, for the female trans- mits more cytoplasm than the male. Certainly the hybrid between a large hen and a bantam cock starts life on a very different plane of size from the hybrid between a bantam hen and a large cock. A writer in Wright's Poultry Book (1902) says in respect to breeding Houdans that the male bird is more responsible for the outside qualities — color, size of crest, beard, tail carriage, color of legs, and so on. The hen determines laying qualities and general size. I have made only one extensive experiment on this matter. I crossed a single-comb White lyCghorn bantam and a Dark Brahma both ways. The offspring of the Dark Brahma hen (weight. 1,300 grams) are a little heavier than those of the White I^eghorn bantam hen (weight, 700 grams). Two males descended from the one and the other mother, respectively, weighed at 3^ months 720 and 550 grams. The average of three pullets from the Dark Brahma at 3 months 22 days is 655 grams ; of three pullets from the White lyCghorn at 3 months 23 days is 626 grams. The proportional differ- ence in the weight of the young of about 3 to 4 months is less than that of their parents, but is in the same sense. The booting of the offspring of the White I,eghorn hen is much reduced as compared with the booting of the offspring of the Dark Brahma hen, the father in the first cross not differing from the mother in the second cross in its heavy booting. In plumage color the 19 offspring of the White Leg- horn female were all white except four. Of the 19 offspring of the Dark Brahma female, only six were white, the others resembling the Dark Brahma. Thus we see that in these three characters of weight, booting, and plumage color the offspring tended to ' ' take after ' ' the mother. INHERITANCE OF SEXUALITY DIMORPHIC CHA RACTERISTICS AND SEXUAL DIMORPHISM IN THE HYBRIDS. Most species of vertebrates exhibit certain characteristics peculiar to one or the other sex, and it is well known that, for example, a female peculiarity can be transmitted through a son to a granddaughter. Thus the good milking quality of a cow is transmitted through her son to his daughters. Whenever femaleness crops out in the history of the germ plasm the good milking quality, or whatever other quality it may be, also appears. The inheritance of dimorphic characters is most strikingly seen in hybridization. Thus I crossed a male Tosa fowl (which has self-colored feathers 1 with a white Cochin.-'^ The male hybrids are barred with white, but the female hybrids closely resemble in color the female Tosa fowl in having white * See Series IX. 94 INHERITANCE IN POULTRY. shafting on the contour feathers, although the white shafting is much broadened. When the barred male and broad-shafted females of this first hvbrid generation were crossed the pure plumage of the Tosa fowl tends to reappear. The males have contour feathers without white and with much red ; the females have the shafted feather without any red. With maleness or femaleness go the proper secondar}' attributes. What is true of the Tosa fowl is true generally, and there is much oppor- tunity to test this matter in poultry, for sexual dimorphism is widespread. In all "dark" or "partridge," silvered, and golden races as found iu Brahmas, Cochins, Wyandottes, Dorkings, Hamburghs, Games, and Oriental fowl, the plumage of the two sexes is conspicuously different ; and to pro- duce sexual dimorphism in a race that is without it the use of one male bird of a dimorphic race may suffice. Again, in the male, comb and wattles are generally larger than in the female. The rose comb of the male becomes often a modified pea comb in the female. The simple comb of the Minorca, Spanish, and Dorking fowl is erect in the male, drooping to one side (equally to the right and left side) in the female. The form of the hackle and saddle feathers constitutes one of the most constant differences between the two sexes. These are long, narrow, and pointed in the male ; short, broad, and rounded in the female. The tail feathers differ similarly. The sickle feathers and those of the middle row especially continue to grow in the male long after their growth has ceased in the female. Similarly the crest feathers of Polish and Houdans grow longer in the male than in the female, but on account of theirgreater breadth in the female her crest appears larger and fuller. Lastly, the greater development of the spurs in the male over six to eight months old is a well-marked dimorphic character. Of these characters I have paid most attention to plumage and skin color, and will take up in review the results gained in crossing dimorphic species. Black Minorca and Dark Brahma. — The male Dark Brahma has white- laced hackles and black, white, and red wing bars. In the female the lacing on the hackles is less conspicuous, and there are no wing bars or bows. Red is wholly absent.- All hybrids are prevailingly black. All males, how- ever, show a more or less prominent wing-bar formed of black, straw, and red colored feathers. No females show any trace of a wing bar unless it be a slight iridescence in the wing coverts. White Leghorn and Dark Brahma. — The male hybrids are typically white, with some red on wing coverts. Apart from some black individuals, the female hybrids are either white, with some buff on wing, or else they resem- ble the female Dark Brahma, having the penciling modified into mossiness. There is no well-defined wing-bar, but the middle wing is suffused with red. White Leghorn a?id Hotidayi. — Neither of these races exhibits a marked dimorphism in plumage color. Nevertheless, the coloration of the hybrids is dissimilar in the two sexes, the males being of a much purer white than GENERAL TOPICS IN INHERITANCE. 95 the females (p. 21), and this is true not only in the first generation, but also in the extracted whites of a later generation. White LegJiorn and Rose-comb Black Minorca. — In the first generation the male hybrids were almost without exception pure white ; the female hybrids invariably show some black-speckled feathers. Tosa fozvl and White Cochin Bantain. — In the first hybrid generation, as stated, the males had all feathers of male Tosa coloration, but barred with white. The females had the Tosa hen coloration, but with shafting broad- ened. Here each sex inherits the corresponding characteristic plumage of the Tosa fowl modified by the white of the Cochin, but in different fashion for each sex. Barring or cuckoo marking seems, indeed, a prevailingly male characteristic. Hurst (1905, p. 133), in crosses of White Leghorn and Houdan, got, in addition to white hybrids, 11 dark birds ; of these the 6 pullets were black ; the 5 cockerels were barred. In the F, generation I obtained extracted pure (?) male and female Tosa- fowl plumage as well as pure whites (p. 49). Dark Brahma {female) and Tosa Jowl {tnale). — Here both races are dimorphic. The female hybrids closely resembled in coloration the female Tosa fowl, except that the contour feathers were penciled as in the Brahma. The male hybrids closely resembled in coloration the male Dark Brahma, except that much more red and less white appeared on the wing bars and wing bows. Both sexes inherit some qualities from the corresponding sex of each of the parent species. Again, the males have a yellow foot like their mother, whereas the females have a willow foot like their father. The hybrids of either sex inherit foot-color from the opposite sex of their parents (p. 54). TRANSFER OFSEXUAIXY DIMORPHIC CHAR ACTKRISTICS FROM ONE SEX TO THE OTHER. Secondary sexual characters, such as have been referred to in the last sec- tion, seem indissolubly associated with their corresponding sex. The reason for such an association is obscure, but it is known that it is not due to the absence in the protoplasm of the characteristics of the opposite sex, for these may develop in the individual when the germ glands are removed. The germ glands, then, control the latency of the one set of characters and the patency of the other set. In poultry the removal of the sex glands from a young cock, in the process of caponizing, results in loss of the crowing instinct and failure of comb, wattles, spurs, hackle, saddle, and sickles to acquire the size characteristic of a cock. If in the fowl the germ glands fail to develop, the secondary sex characters are ambiguous. Despite this apparently physiological dependence of secondary characters in the germ gland, it seems improbable that the association is a necessary one. Almost all characters can be dissociated ; wh}'' not also sex and second- ary characters ? There is reason to think much can be done in this way, because something has alreadj^ been accomplished. For example, the cereb- 96 INHKRITANCK IN POULTRY. ral hernia which now is found equally in both sexes of the Polish fowl was formerly a female secondary sexual characteristic. Bechstein (1793) states that he never observed the cranial dome in male Polish fowl. Blumenbach (1813), who made numerous dissections of the cranium of this fowl, states "of this deformit}' very slight traces indeed are found in the cocks, and these but seldom. ' ' * Consequently it must be concluded that the female secondary characteristic of cerebral hernia has been gradually transferred to the male sex also. A case of which the history is known even more defi- nitely is that of the vSebright Bantam. This bird is characterized by the fact that in the male the hackle, saddle, and sickle feathers are of the same form as in the female ; consequently the tail is short and truncate as in a hen. Here, apparently, female characteristics have become attached to a male. Fortunately we have the history of the race from the mouth of the son of the maker, Sir Thomas Sebright. Dr. Horner, who obtained the state- ment from Sebright, published it in Tegetmeier's Poultry Book (1868, pp. 241, 242). It was about the year 1800 that the late Sir John Sebright began to fashion the Sebright Bantam. The first cross was between a common I^antamf and the Polish fowl.| The chickens resulting from this alliance were bred in-and-in until the required markings and size were secured. Sir John then accidentally found a hen-tailed Bantam cock in the country where he was traveling. This short-tailed bird he in-bred with his newly manufactured Bantams, thereby giving their progeny the present form of the square tail. The essential characteristic of the race was thus gained from a mutative modification of a polymorphic characteristic. § In my own experiments I have hardl}'' proceeded far enough to get results ; yet already evidence of transference of color characteristics from one sex to the other is appearing. Thus in the .second hj^brid generation of the Cochin X Tosa cross at least one bird (No. 659 9 ) kas hackles of a plain buff color like those of the male Tosa fowl, and entirely unlike the hackles of the female Tosa fowl or the female of the dark variety of the Cochin. Again, the female hybrids between the Dark Brahma hen and White Leghorn cock have much red on the wing coverts. This is foreign to the Dark Brahma hen, and must, so far as I can see, have been derived from the red on the wings of the male Tosa fowl. Finally, two male hybrids between the Tosa fowl and Dark Brahma show the feminine shafting. Experiments in con- tinuance of this investigation are, naturally, in progress. * Translation quoted by Tegetmeier, 1867, p. 173. t Doubtless Game Bantam is here meant. t The Golden Spangled Polish are undoubtedly referred to, whence the spangling of the feather was obtained. The combination gave the small size and gold-spangled plumage. § As might be anticipated from the notoriously sterile quality of hen-feathered cocks, Sebright Bantams are inclined to be sterile, and one is advised not to try to breed from the best show stock. 7. e., cocks with the shortest tails (Wright, 1902, p. 598). GENERAL TOPICS IN INHERITANCE. 97 SEX IN HYBRIDS. There is a widely held and frequently expressed opinion that hybrids show an excessive proportion of males. Bate.sou and Saunders (1902, p. 139) probably have this in mind in their statement — " the statistical distribution of sex araonof first crosses shows great departure from the normal propor- tions." I have therefore been interested to tabulate the sex proportions in my hybrids. Without giving the full table, I may state that the totals are : Males, 204 ; females, 173 ; sex undetermined, 573. There is here an excess of males ; but in view of the large early death rate, this may well be due to a difference in the death rate of the two sexes. Taking the different series of hybrids separately, most of them gave an approximation to equality of the sexes. One of the most striking departures is the series of Dark Brahma (121 $) X Tosa (8a ^) hybrids. Of 22 individuals that developed to 18 days in the incubator, all but one grew to maturity. Of these 21, 16 are males and 5 females. The first egg laid by the Dark Brahma after she was put with the Tosa fowl developed into a female ; the next nine that hatched were males ; also her last six young were males. The exceptions to the law of equality of sexes in hybrid offspring are thus individual and not of general significance. CORRELATION OF CHARACTERISTICS. Every taxonomic description testifies to the fact that a certain set of characteristics is usuall)'- found associated in each species or variety. The prevailing theory has been that this association is a necessary one, maintained because all the characters are necessary to the success of the species in its relations to external environment, or else that they were physiologically inter- dependent. Modern work in hybridizing is establishing the fact that few of the specific characteristics are interdependent. Their association is, so far as interaction goes, mostly accidental. Thus in my experiments with poultry I have merely reached the same conclusions as have been gained by Johannsen (1899, p. 185), de Vries (1903, p. 494), and indeed all recent workers. I find, namely, that of the scores of evident external characteristics of poultry that are inherited in alternative fashion scarcely two can be found that are always associated. The most striking exception is the association of high nostril and absence of single comb. What, then, is the meaning of correlation in nature? Clearly it is only rarely due to physiological interdependence. It may often be due to an unrelated association of characters independently advantageous to the organ- ism. It is doubtless due to an accidental association of characters brought into the race by successive mutations or by hybridizations and never disturbed, because not prejudicial to the well-being of the species. 98 INHERITANCE IN POULTRY. THE MrTATION THEORY IN ITS RELATION TO THE ORIGIN OF DOMESTICATED ANIMALS. While the mutation theorj' of de Vries has received widespread adherence among botanists, many students of animals, and especially of domesticated races, have appeared as its opponents. Foremost among these are Professors Keller, of Zurich (1905), and Plate (1905), of Berlin. I think that the essence of the mutation theory is too little apprehended. It rests on the funda- mental theory of heritable unit characters and assumes their very limited mutability. It recognizes the important results wrought by artificial selec- tion, but considers them as arising from two processes — first, the selection of minute favorable variations of the fluctuating sort, and, secondlj^ the preser- vation of new unit characters suddenly appearing. Such unit characters can usually be not only maintained but much improved by subsequent selective breeding. Now, it is true that breeders nowadays do not regularly wait for favorable qualities to crop out. The process is too slow, uncertain, and expensive. If one had scores of thousands of individuals, desired mutations would come more frequently ; but even then they would rarely be of a desirable sort. Every breeder can, on the other hand, improve any characteristic by selec- tion, and that is for the most part the only method of improving a quality that is open to him. Of course he can make new combinations of qualities by crossing, but this does not, typically, result in new qualities. The question of the permanence of the improvement wrought by selection of minute variations is the first point of difference between de Vries and Keller. De Vries asserts that such improvement persists only so long as selection is maintained. Keller adduces some interesting cases on the other side, and the cogency of some of his evidence must be admitted. He traces the gradual evolution in Egypt of long lop-eared hounds from straight-eared ones. Ear length in rabbits, as Castle (1905, pp. 125-126) has shown, is not a unit character ; at least, it blends in hybridization and con.sequently exhibits any desired intermediate condition. The same is probably due for dogs ; consequently this character may well have arisen by summation of minute variations. Yet Keller goes on to show the long-eared condition has per- sisted in central Africa, where selective breeding no longer occurs. Hence one characteristic originated by selection of fluctuations has not retrogressed on removal of selection. The preceding method of proof is not, however, of general validity. Evidence that a characteristic arisen in domestication does not disappear when the race becomes feral again is not evidence against the permanence of fluctuations unless it is also proven that the characteristic arose by selec- tion of fluctuations. This is usually not the case. The instance of long ears would seem to be peculiar. Some of the other examples offered by Keller of persistence of characteristics despite discontinuance of selection avail little, since the precise origin of the unit characters concerned is un- GENERAL TOPICS IN INHERITANCE. 99 known. If, unconsciously or not, a unit character arising as a sport has been preserved under domestication, it will persist even though the race bearing it become feral. Positive support for the mutation theory is gained from a consideration of the characteristics of poultry. Our study has shown them to be, for the most part, of the order of integral unit characters. As such they could hardly have been " gradually built up." Being indivisible they must have appeared at once, roughly in their present form. The very existence of unit characters is proof of the mutation theory. That many characteristics of organisms have not been built up, but have suddenly appeared complete, may be inferred from peculiarities of the char- acters other than their integral nature. For, first, not all kinds of charac- teristics have been evolved in domestic poultry, but for the most part only such as occur elsewhere among wild races. Thus, for example, booted feet, as found in the grouse ; crest on head, as seen in the umbrella bird {Ccphahp- terus), and long tail, as seen also in the widow bird {Chera). Secondly, many of the characteristics of domestic poultry are of the order of mutations in so far as they are almost pathological, e. {r., taiUessness, rose comb, silky and frizzled feathers, cerebral hernia, polydactyl feet, albinism. These char- acters, cropping out in the sporting organism and not being prejudicial to its well-being, have been preserved by the fancier ; they doubtless arose sud- denly, as we find arising suddenly to-day other characters, which we discard because incompatible with a healthy stock — such as featherlessness, cross-bill, and imperfect development of toes. If these characteristics appeared sud- denly and not by being "built up," as we know is the case, then so, doubt- less, have others. The evidence that many, if not most, characteristics of poultry have arisen suddenly, without having been sought and laboriously built up by man, is convincing, and there can hardly be any escape from the conclusion that here evolution has bee.i largely, though not exclusively, by mutation. lOO INHKRITANCE IN POULTRY. E. SUMMARY OF CONCLUSIONS. (i) Poultry exhibit nnnieroiis unit characteristics which do not blend in hybridization, but are inherited in alternative fashion. The unit characters are not immutable things in hybrids, but subject to modification — perhaps permanent — by interaction of the alternative characters. (2) Although the great majority of characteristics of poultry are inherited alternatively, yet a few cases of color characters show a particulate inher- itance. The comparative rarity of blending of characters makes it easier to see how new characters will not be ' ' swamped by intercrossing with the parent form " (page 82). (3) Specific and varietal characteristics in de Vries's sense are not inherited in a markedly different fashion, although in two cases progressive variants do not Mendelize typically. (4) The patent characteristic is usualh^ dominant over its latent allelo- morph. (5) Old and new characteristics are equall}^ dominant. (6) Dominance and recessiveness of characteristics are not always accom- paniments of their segregation in the germ cells ; both, moreover, are fre- quently incomplete. (7) Dominance is usually , but not always, independent of the races crossed. (8) Prepotency is as truly important in inheritance as dominance. (9) Many first hybrids exhibit special forms, due to the interaction of the two allelomorphs. These may become fixed as new characteristics. (10) Reversion is being explained bj^ the persistence in a "latent" condition of the latent character. (11) An adequate theory of gametic purity has not only to explain the simple Mendelian formula, but also the facts of imperfect dominance, im- purity of extracted forms, latency and atavism, and occasional particulate inheritance. (12) Reciprocal crosses exhibit differences due to the fact that the father and the mother transmit different kinds of characteristics. (13) When the parent races are dimorphic each .sex in the hybrids exhibits the respective sex characteristic of both of the species. In many cases a new form of sexual dimorphism appears in the hybrids. (14) Certain characteristics of one sex may become transferred to the other by hybridization, owing to lack of complete correlation between primary and secondary sex characters. (15) The proportion of the two sexes in hybrids is normal. (16) With few exceptions, correlated characteristics easily separate as a result of hybridization so that any conceivable combination may be effected. Carnegie Institution, Station for Experimental Evolution, Cold Spring Harbor, February 12, igo6. LITERATURE CITED. lOI F. LITERATURE CITED. American Poultry Association. 1905. The American standard of perfection. Illustrated. A complete description of all recognized varieties of fowls. Published by Amer. Poultry Assoc. 1905. 299 pp. BaIvDamus, a. C. E. 1896. Illustrirtes Handbuch der Federviehzucht. Erster Band : Die Hiihnervogel. 3 Aufl. bearbeitet von O. Griinhaldt. Dresden : Schonfeld. 1896. xvi + 476 pp., 102 figs. BaTeson, W., and Saunders, Miss E. R. 1902. Report I to the Evolution Committee of the Royal Society. London : Harri- son. 160 pp. BaTESON, W., and PunneTT. 1905. Experimental studies in the physiology of heredity— Poultry. Report II to the Evolution Committee of the Royal Society, pp. 99-131. Bechstein, J. M. 1793- Gemeinniitz. Naturgesch. Deutschlands. Bd. 3. Sumpf- u. Hausvogel. Leip- zig : Vogel [teste Darwin, 1876]. BI^UMENBACH, J. F. 1805. Handbuch der vergleichenden Anatomie. Gottingen : Dietrich [teste Darwin, 1876]. 1813. De anomalis et vitiosis quibusdam nisis formativi aberrationibus Cum tab. 4°. Gottingen : Dietrich. BOREi-Li, P. 1670. Historiarum et Observationum medicophysicarum Centuriae iv. Francofurti. BOVERI, T. 1902. Ueber mehrpolige Mitosen als Mittel zur Aualj'se des Zellkerns. Verh. d. phys.- med. Ges. zu Wiirzburg. N. F. Bd. xxxv. Cannon, W. A. 1902. A cytological basis for the Mendelian laws. Uull. Torrey Bot. Club. Vol. 29. Castile, W. E., and Allen, G. M. 1903. The heredity of albinism. Proc. Amer. Acad, of Arts and Sciences, xxxvii, 603-622. April. Castle, W. E. 1903°. The heredity of "Angora" coat in mammals. Science, n. s., xvni, 760, 761. Dec. II. 1905. Heredity of coat characters in guinea-pigs and rabbits. Publication No. 23, Carnegie Institution of Washington. Papers of Station for Experimental Evolution No. 2. 78 pp., 6 plates. Feb. Chamberlain, B. H. 1900. Note on a long-tailed breed of fowls in Tosa. Trans. Asiatic Soc. Japan, XXVII. Clayton, J. 1693. Philosophical Transactions of the Royal Society of London, 1693, p. 992. CORRENS, C. 1900. Mendel's Regel iiber der Verhalten der Nachkoramenschaft der Rassenbas- tarde. Ber. d. deut. bot. Ges , xviii, 158-168. Sitzung von 27 Apr. 1905". Gregor Mendel's Briefe an Carl Niigeli, 1866-1873. Ein Nachtrag zu den veroffentlichten Bastardierungsversuchen Mendels. Abh. math.-phys. Kl. k. siichs. Ges. d. Wiss., xxix, No. 3, pp. 189-265. 1905''. Einige Bastardierungsversuche mit anomalen Sippen und ihre allgemeinen Ergebnisse. Jalirb. fiir wiss., Bot. XM, Hft. -^, pp. 458-484. Taf. v [April]. 1905". Zur Kenntnis der scheinbar neuen Merkmale der Bastarde. Bericht der Deutschen Botan. Gesell., xxiii, 70-85. 1905'*. Uber Vererbungsgeselze. Berlin: Borntraeger. 43 pp., 4 figs. 2 INHERITANCE IN POULTRY. CUENOT, L. J903. L'h^r^dite de la pigmentation chez les souris (2°* note). Arch, de zooL exper. et gen. (4) 1. Notes et rev., pp. xxxiii-xli. CUNMNGHAM, J. F. 1903. Observations and experiments on Japanese long-tailed fowls. Prop. Zool. Soc. London, 1903, I, pp. 227-250. Darwin, C. 1876. The variation of animals and plants under domestication. Second edition, revised. Vols, i, 11. New York : D. Appleton & Co. [References made, in brackets, to pages of reprint by D. Appleton " Fourth thousand, 1894."] Davenport, C B. 1904. Wonder horses and Mendelism. Science, xix, 151-152. Jan. 22. DE Vries, II. 1889. Intracellulare Pangenesis. Jena : Fischer. 212 pp. 1900. Sur la loi de disjonction des hybrides. Conipt. Rend, de I'Acad. des Sci. Paris. 26 mars. 1902. Die Mutationstheorie. Versuche und Beobachtungen iiber die Entstehung der Arteu im Pflanzenreich, Zweiter Band. 1. Lieferung, pp. 1-240. 1903. Die Mutationstheorie. 11. Bd. 2. Lief, pp. 241-496. 1905. Species and varieties : Their origin by mutation. Ed. by D. T. MacDougal. Chicago : Open Court Publishing Co. 1905. x\iii-)- 847 pp. DURIGEN, B. 1886. DieGefliigelzucht nach ihrem jetzigen rationellenStandpunkt. Berlin: Pare}'. 880 pp. 80 Taf u. loi fig. in text. FOCKE, W. O. 1881. Die Pflanzen-Mischlinge. Ein Beitrag zur Biologic der Gewachse. Berlin : Borntraeger. iv + 567. Galton, F. 1883. Inquiries into human faculty. Loudon : Macniillan. 1889. Natural inheritance. New York and London, ix + 259 pp. Haacke, W. 1893. Gestaltung und Vererbung. Eine Entwickelungsmechauik der Organi^men. Leipzig: T. O. Weigel Nachfolger (Tauchnitz). vi + 337. Hacker, v. 1904. Bastardirung und Geschlechtszellenbildung. Zool. Jahrb., Suppl. yii (Fest- schrift fiir Weismann). Hagenbach, E. 1839. Untersuchungen iiber den Hirn-und Schadelbau der sogenannten Hollen- hiihner. Archiv. fiir Anat. Physiol, u. wiss. Med. (Miiller) Jg. 1859, pp. 311-331, Taf. XVI. Hurst, C. C. 1904. Mendel's discoveries in heredity. Trans. Leicester Literary and Philos. Soc, VIII, pp. 1 21-134. June. 1905. Experiments with poultry. In Report II to the Evolution Committee of the Royal Society (by Bateson f/ «/.). London : Harrison. 154 pp. Johannskn, W. 1899. Sur la variability de I'orge consider^e au point de vue special de la relation du poids des grains a leur teneur en matieres azotiques. C. R. trav. de Lab. de Carlsberg. iv. Heft., 4, pp. 122-192. KELI.ER, C. 1905. Die Mutationstheorie von de Vries im Lichte der Haustier-Ge.'^chichte. Arch, fiir Rassen-und Gesellschafts-Eiologie. 11 Jg., 1. Heft. , pp. 1-19. Feb. Langkavel, B. 1886. Hiihner mit sechs Zehen. Der Zoologische Garten, xxvii, p. 35. Jan. Lucas, P. 1847. Traits philosophique et physiologique de I'her^dite naturelle dans 6tats de sant^ et de mala die du systeme nerveux., etc. Tom. i. Paris : J. B. Bail- Here. 1847. 24 -t- 626 pp. 1850. [Same title.] Tom. 11. Paris : J. B. Bailliere. 1850. 936 pp. WTKRATURE CITED. IO3 McClung, C. K. 1905. The chromosome complex of orthopteran spermatocytes. Biol. Bull., vol. ix, No. 5, Oct., pp. 304-340. McGrew, T. F. 1901. American breeds of fowls. I. The Plymouth Rock. Bull. No. 29, Bureau of Animal Industry, U. S. Dept. Agr. Washington : Government Printing Office. 1901*. American breeds of fowls. II. The Wyandotte. Bull. No. 31, Bureau of Animal Industry, U. S. Dept. Agr. Washington : Government Printing Office. 30 pp. 1904. The Shanghai or Cochin fowl. In Weir-Johnson-Brown Poultry Book. pp. 523-558. Mendel, G. 1866. Versuche liber Pflanzen-Hybriden. Verhandlungen des naturforschen Vereines in Briinn. Bd. iv, 47 pp. Morgan, T. H. 1905. The assumed purity of the germ cells in Mendelian results. Science, vol. xxii, No. 574, pp. 877-879. Dec. 29. Nageli, C. v. 1884. Mechanisch-physiologische Theorie der Abstammungslehre. 1898. A mechanico- physiological theory of organic evolution. Summary. Transla- tion by F. A. Waugh. Chicago : Open Court Publishing Co. 53 pp. Petersen, C. E. 1905. The Houdan. In Weir-Johnson-Brown Poultry Book, 1904-05 {g. v.). Plate. L. 1905. Die Mutationstheorie im Lichte zoologischer Tatsachen. C. R. du 6"" Congr^s intern, de Zool. Berne. 1904. 203-212. May 25. Romanes, G. J. 1901. Darwiu and after Darwin. I. The Darwinian theory (third edition). Chicago : Open Court Publishing Co. Standfuss, M. 1896. Handbuch derpalaarktischenGross-Schmetterlingefiir Forscherund Sammler. Jena : Fischer, xii + 392. 8 Taf, Sutton, W. S. 1902. On the morphology of the chromosome group in Brachystola magna. Biol. Bull., IV, No. I, pp. 24-39. Dec. 1903. The chromosomes in heredity. Biol. Bull., toI iv, No. 5, pp. 231-251. April. TegetmeiER, W. B. 1856. On the remarkable peculiarities existing in the skulls of the feather-crested variety of the domestic fowl, now known as the Polish. Proc. Zool. Soc. Lond., 1856, pp. 366-368. Figures. 1867. The poultry book, etc. London : Routledge. viii + 356 pp., 30 colored plates, 36 imcolored figures. Thorndike, E. L. 1905. Measurements of twins. Archives of philosophy, psychology, and scientific methods. No. i, Sept. 64 pp. Tschermak, E. 1904. Weitere Kreuzuiigsstudieu an Erbseu, Levkojen und Bohnen. Zeitschr. fiir das landwirths. Versuchs\vesen in Oesterr. 1904. 106 pp. Weir, H. ; Johnson, W. G., and Brown, G. O. 1904-05. The poultry book. 3 vols. New York : Donbleday, Page & Co. xxii -f 13 II pp. Weismann, a. 1904. The evolution theory. Trans, by J. A. Thomson and M. R. Thomson. 2 vols. Loudon : Arnold, xvi -f 416 -f- 405 pp. Woods, F. A. 1902-03. Mental and moral heredity in royalty. Popular Science Monthly. .\ug. 1902-April, 1903. Wright, L. 1902. The new book of poultry. London, etc. : Cassell & Co. viii ^- 600 pp. Wyckofe, E. G. 1904. The Leghorns. In Weir-Johnson-Brown Poultry Book. 1904-05 {q. v.). ZlEGEER, H. E. 1905. Die Vererbungslehre in der Biologic. Jena : Fischer. 74 pp. 2 Taf. I04 INHERITANCE IN POULTRY. EXPLANATION OF PLATES. PI.ATE I. Pig. I. — White Crested Black Polish, 9 5- One of the females crossed with the Single- combed Black Minorca ((/. fig. 3) to produce the female hybrid shown in fig. 5. (H.A. H.) Fig. 2. — White Crested Black Polish, r^ 30. To show the male type of the Polish race, which, when crossed with the Minorca (fig. 4), produces male hybrids like fig. 6. (H. A. H.) Fig. 3. — Single Comb Black Minorca, 9 13- The mother of the hybrids, Minorca X Polish, represented by fig. 5. Fig. 4. — Single Comb Black Minorca, cj' 12. The father of various Polish X Minorca crosses, of which a male is represented in fig. 6. Fig. 5. — First Hybrid between Polish and Minorca, pullet. Compare the females of the parental races, figs, i and 3. Fig. 6. — First Hybrid between Polish and Minorca, cockerel. Compare the males of the parental races, figs. 2 and 4. The fig^ures marked H. A. H. are from photographs made by Mr. H. A. Hackett. PLATE I. I06 INHERITANCE IN POULTRY. Pirate II. Fig. 7. — The head of a Polish fowl, (^ 3, with skin on left half of head dissected away. Shows cerebral heruia, and the relation to it of the thick skin and crest feathers lying above. Note also the cultninal fold, high nostril, and rudi- mentary comb. The latter lies at the base of the comb and shows as a mottled area against the deep black of the anterior crest feathers. (H. A. H.) Fig. 8. — Head of a hybrid, (^ 50, between Minorca 9 13 (iig. 3) and Polish ^f 3 (fig. 7). Shows the Y-shaped comb Ijing in front of the crest. The comb is double behind, single in front. Fig. 9. — Head of a Minorca X Polish hybrid of the second generation. The son of such a pair as are represented in figs. 5 and 6. Note the reappearance of a large crest, high nostril, and rudimentary comb. (H. A. H.) Fig. 10. — Side of cranium of Polish fowl. Shows bony dome covering cerebral hernia. PLATE II \'J I08 INHERITANCE IN POULTRY. Plate ill. Fig. ir. — Head of hybrid of the second generation, Minorca X Polish, 371 (j" — the son of such a pair as are represented in figs. 5 and 6. Note the absence (imper- fect) of crest, the high nostril, and the rudimentary comb. (H. A. H.) Fig. 12. — Head of Houdan {(^ 9). Shows crest, high nostril, and rudimentary comb. Fig. 13. — Foot of Houdan {(^ 9). Shows the two toes (in place of one) situated imme- diately below the spur. Fig. 14. — Head of second generation White Leghorn y Houdan hybrid, its father being like fig. 17. Shows the occurrence of both cerebral hernia and single comb on the same individual, (H. A. H.) PLATE li 11 ^,».. ^^ I ^ ^ ^^. * :^ :^' no INHERITANCE IN POULTRY, PI.ATE IV. Fig. 15. — Single Comb White Leghorn (cf 74). Note high single comb, plain head, clean feet with four toes, and white plumage. (H. A. H.) Fig. 16. — Houdan ( 9 8). Note crest, high nostril, rudimentary comb, mottled plumage, and mufT and beard. Fig. 17. — First hybrid {^^f 87) between White Leghorn and Houdan. Note crest, Y comb, white plumage, muff, and double toe behind on left foot. PLATE IV. sS^i*:- % 17 •4 > 15 .j^f^ V \ ^^\ 112 INHERITANCE IN POULTRY, Plate V. Fig. i8. — Dark Brahma hen (121), Note uniformity of plumage coloration, except that hackles are laced ^nih. wbit?, and'^\ing coverts, back, and breast are penciled. Comb of pea type. Feet booted. Vulture hock. (H. A. H. ) Fig. 19. — Dark Brahma cock (122). Isote laced hackles and saddles, prominent white wing bow, pea comb, and booted feet. PLATE V. 114 INHERITANCE IN POULTRY. Plate VI. Fig. 20. — First generation hybrid, (^ 607, between Black Minorca (Sg. 6) and Dark Brahma (fig. 19). Note prevailingly black plumage, with trace of white wing bow, irregular pea comb, slightly booted feet, and absence of vulture hock and of lacing on hackles. (H. A. H.) Fig. 21. — First generation hybrid, ^f' 603, between Black Minorca and Dark Brahma. Brother to fig. 20. Note almost complete absence of white wing bow, but presence of white lacing on hackles. Note also high, though pea, comb, and long tail. Form of trunk like Dark Brahma, neck like Minorca, (H. A. H.) PLATE VI. Il6 INHERITANCE IN POUI^TRY. Pl,ATE VII, Fig. 22. — First generation hybritl, Q 387, between White Leghorn Bantam (resembling fig. 15) and Dark Brahma (fig. 18). Note the new type of plumage colora- tion. Hackles broadly laced as in male, rest of plumage mottled, with much red. Booting rudimentary. (H. A. H. ) Fig. 23. — First generation hybrid, 9 395, between White Leghorn and Dark Brahma. Shows reappearance of the Dark Brahma 9 tj'pe of coloration. (H. A. H.) PLATE VII '^■ 5n //»'*;?.:; <^> — • > 22 ^^^ ''^mm IlS INHERITANCE IN POULTRY. PtATE VIII. Fig. 24. — First generation hybrid, cf 270, between White Leghorn and Dark Brahma. Shows the pure white t3'pe, /. ^., dominance of White Leghorn coloration. Note slight booting, absence of vulture hock and the erect, Leghorn tail. (H. A. H.) Fig. 25. — First generation hybrid, rf' 409 A, between White Leghorn and Dark Brahma. Shows the type with red on the wing coverts. (H. A. H.) PLATE VIII ^^r\ 'JT-" 24 \ ■'^' ^ I20 INHERITANCE IN POULTRY. Plate IX. Pig. 26. — Black Cochin Bantam, 9 129. Shows short tail and heavily booted feet. The mother of the barred bird, fig. 27. (H. A. H.) Fig. 27. — First hybrid, rf 365, between Black Cochin Bantam (fig. 26) and White Leghorn {cf. fig. 15). Note barred plumage coloration, red earlobe, and booted feet. (H. A. H.) Pig. 28.— Bufif Cochin Bantam, cT 545- ^'^''^ short tail, heavily booted feet, red earlobe, and single comb. PLATE IX. 122 INHERITANCE IN POULTRY. Plate X. Fig. 29. — Tosa fowl, (^ ia, imported from Japan. Long tail feathers had been recently pulled out. Fig. 30. — Tosa fowl, 9 2A, imported from Japan. Note the light shafting. Fig. 31. — Tosa fowl, cf 3A, "Admiral Togo," son of ia and 2A. Photographed Septem- ber 7, 1905. Note length of tail. (H. A. H.) Fig. 32. — White Cochin Bantam, 9 iS^- This bird was crossed with fig. 29 and gave hybrids represented on plate XI. PLATE X. 32 124 INHERITANCE IN POULTRY, PI.ATE XI. Fig. 33. First generation hybrid, $ 58, between White Cochin (fig. 32) and Tosa fowl (fig. 29). Note a slight broadening of shaft stripe as compared with female Tosa fowl. Fig. 34, — First hybrid, rf 53, between White Cochin and Tosa fowl (fig. 29). Note white barring on feathers, and long tail. (H. A. H.) Fig. 35. — First hybrid, cf 95, between White Cochin and Tosa fowl, younger brother to fig. 34. Note barring and growth of saddle and tail feathers. (H. A. H.) Fig. 36. — Second generation hybrid, 9 312, between White Cochin and Tosa fowl. Note pure white plumage color, like Cochin grandmother, fig. 32, combined with long tail of Tosa, fig. 30. (H. A. H.) PLATE XI ^ 36 "-^ 35 126 INHERITANCE IN POULTRY. Plate XII. Fig. 37. — Plumage chart of F, (White Cochiu X Tosa), 9 58. at about 5 months, pi, to of head ; 2, hackle ; 3, middle of back ; 4, throat ; 5, breast ; 6, middle tail ; 7, saddle ; 8, wing, secondary. Fig. 37a.— Plumage chart of F, (White Cochin X Tosa), cf 53, at about 5 months. Signifi- cation of figures same as in fig. 37. Shows barring of feathers. Fig. 38. — Second hybrid generation (White Cochin X Tosa), J^ 315. Note reappearance of pure white like Cochin grandmother (fig. 32, plate X) ; form intermediate, feet booted. (H. A. H.) 37= 37 128 INHERITANCE IN POX"'LTRY. Plate xill. Fig. 39. — Jungle fowl, 9 2. Taken after death to show shafting on breast, nape, back, and wing coverts. Fig. 40. — First ^reneration hybrid, (^ 358, between Dark Brahma (fig. 19, plate V) and Tosa fowl (fig. 29). Note white laced hackles and saddles, vulture hock, boot, and pea comb of Brahma and white earlobe and elongated head of Tosa. (H. A. H.) PLATE XII 39 /v^(@> 40 130 INHERITANCE IN POULTRY. Plate XIV. Pic. 41. — Frizzle fowl, cT ^5- Note rose comb and feathers that turn forward, forming a ruff on the neck. On the exposed vanes of the upper secondaries the twist- ing of the barbs may be seen. Fig. 42. — Frizzle fowl, 9 i8a. Note extreme curling of feathers, the absence of barbs on part of the secondaries, leaving the shaft quite naked, and the absence of plumage on the back of the head. Pig. 43. — Silky fowl, (^f 21A. Note single comb, small crest, the downy condition of the contour feathers, and the elongated aud disconnected barbs of the wing secondaries and tail feathers. Pig. 44. — First hybrid between Frizzle and Silky, cf 219. Note the white plumage, rose comb, trace of crest, frizzled feathers (ruff !), and absence of elongated barbs on the wing, secondaries, or other evidence of silkiness. The booted feet and extra toe are derived from the Silky. (H. A. H.) rLA I t AlV. CO I .hf ^^t^m'^ w ^ ■V ~>. 4^ 132 INHERITANCE IN POULTRY. PI.ATE XV. Fig. 45. — Runipless Game, 9 49- Fig. 46. — Ruinpless Game, J"' 117. The hackles and saddles and wing bars are red ; other- wise the plumage is largely black. (H. A. H.) Fig. 47. — First hybrid between White Leghorn {cf. fig. 15, plate IV) and Rumpless Game, (J' 516. Note dominance of white (except for a trace of red on the wing coverts) and long tail. The comb is intermediate between that of a Game and that of a White Leghorn. (H. A. H.) PLATE XV. I?£ •vV % \r^ 47 134 INHERITANCE IN POULTRY. Plate XVI. Fig. 48. — Head of second generation Minorca X Polish hybrid, (J* 474. Shows last term of series, beginning with fig. 50 and passing through fig. 49, of reduction of median component of Y comb, leaving only two papillae remaining. (H. A. H.) Fig. 49. — Head of second generation Minorca X Polish hybrid. Shows middle term of series passing from Y comb to V comb. The median portion of the comb is represented by a carunculated mass at the base of the papillae. (H. A. H.) Fig, 50.— Head of second generation Minorca X Polish hybrid, ^f 259. Shows beginning degeneration of median component of Y comb, which ends in the V comb (fig. 48). (H. A. H.) Fig. 51. — Head of second generation Minorca X Polish hybrid. Shows 2 pairs of papillae, high nostrils and rudimentary crest, indicating that the first two character- istics are independant of the third. H. A. H.) Fig. 52- — Dorsal view of head of (Minorca X Polish) X Minorca hybrid. Shows Y comb in which the median component extends between the arms of the Y, the whole resembling a pea comb. (H. A. H.) TLMl C AVI, O'J D I 136 INHKRITANCE IN POrLTRY. Plate XVII. Fig. 53. — First generation hybrid between Silky and Jungle fowl, (^ 156. Shows domi- nance of Jungle-fowl plumage color and the extra toe and crest of the Silky. (H. A. H.') Fig. 54. — First generation hybrid between White Leghorn and Rose Con.b Black Mi- norca, 9 1 38. One of the two birds that exhibit the blue, Andalusian type of coloration, all others being white. (H. A. H.) Noftb^aroHna State Library Raleigh » 4^ PLATE XVII /■^iTTSwiiii -^v? 53 <^f »N • X: :■; ^ ':^v ^, X 54 -«e^ INHERITANCE IN POULTRY S02776212 c' M, >'^' ^' />; %:4 A v^ (^ r r 0i ^> i'C:,. !^' "5-,?'.. - /';ii 1*1, S h$ S: &■ :^*||