|Fig. 10. A cob of Maize (Zea mays) borne by an F1 plant from the cross round x wrinkled, fertilised with its own pollen, showing the mixture of round (dominant) and wrinkled (recessive) seeds. (From a specimen given by Dr Webber.)
[Figure not reproduced in this version]
Of the various Mendelian experiments this is one of the most demonstrative. Dominance is perfect so far as external observation goes. Correns records a remarkable excess of round seeds as recurring with great constancy in certain families when F1 is self-fertilised (see later).
It often happens that pollen from one variety of maize is blown by the wind to the stigmas of another variety. If this pollen possesses a dominant factor capable of affecting the seed, seeds exhibiting it are formed. If for instance pollen from a round maize is blown on to a wrinkled or sugar-corn, round seeds will be formed among the normally wrinkled seeds. When formerly it was supposed that the endosperm, which contains the reserve materials, was a maternal structure, the change in the seed was regarded as an influence exerted by the embryo on the maternal tissues. The effects of such influences were called by Focke "Xenia." There are a few examples of such influence which may with probability be regarded as genuine* ; but since the discovery of the fact that the endosperm of maize results from a double fertilisation effected by the second nucleus of the
* The phenomena are discussed by Darwin, An. and Plts., ed. II. 1885, I. pp. 428-433 It seems likely that in some of these instances the factor introduced by the pollen-grain can influence or infect tissues in contact with the embryo.
< 031 >
pollen-tube, cases like that of maize are not strictly to be classed as Xenia (see Correns, 58).
34. Glutenous and starchy endosperms. Wheat. Biffen (27)
Professor Biffen's researches respecting these important features are not yet completed. The glutenous, translucent, hard type has definite dominance over the opaque, soft, starchy type.
35. Single flowers usually behave as dominants to doubles, as to Stocks, Primula, &c. In Carnations the doubleness dominates.
The most extensive researches on the genetics of doubleness are those of Miss Saunders in the case of Stocks (Matthiola), R.E.C.. (20-23). The peculiar phenomena discovered are discussed in a separate chapter (q.v.).
36. In Phaseolus hypo-geal cotyledons are dominant to epi-geal. Various intermediates in F2 Tschermak (278, p. 54).
This list and that which follows make no pretension to completeness. Those features are enumerated which either seem of special interest, or have been studied with some thoroughness. Indications respecting many more are to be found in the original papers (see especially for Peas and Phaseolus the writings of Tschermak and Lock ; for Cotton, Balls ; for Oenothera, &c., de Vries, and Macdougal (186) ; for Wheat and Barley, Biffen, and Tschermak ; for Maize, Correns, and Lock ; for various plants, Correns, and de Vries).
In the genus Brassica numerous crosses have been studied by Sutton (262). In his experiments it was found, among other important results, that the bulbing of the Swede, Turnip, and Kohl Rabi disappeared completely in crosses with non-bulbing Kales, and that in F2 imperfect bulbing reappeared. Professor Biffen, who is continuing work on the same lines, tells me that in regard to these and similar characters cultural conditions play a great part, and lead to curious and conflicting results.
< 032 >
Animals. Structural Characters.
A considerable number of diseases and malformations have been shown to behave usually as dominants. A few conditions may be said, more doubtfully, to behave as recessives. The subject of human inheritance is discussed in Chap. XII. Of normal characteristics, eye-colour is the only one yet studied (Hurst, 161)* sufficiently to justify a positive statement as to the existence of a Mendelian system of descent.
37. Absence of horns in polled breeds of Cattle is dominant to the presence of horns (R.E.C. 19 ; Spillman, 246).
In sheep the inheritance of horns is sex-limited (q.v.), and from evidence given me by Mr E. P. Boys-Smith I suspect that this is true in the case of Goats also.
38. There is little doubt that the gait known as “pacing” is recessive to the ordinary trotting gait in the American trotters. Trotters bred together may produce pacers, but hitherto I have found no authentic instance of genuine natural pacers, when mated together, producing trotters. Correspondents have sent me word of several apparent exceptions to this rule, but all on inquiry have proved to be erroneous. In the pacing gait the two legs of the same side of the body are moved together or nearly so, while in trotting the foreleg of one side moves almost with the hind leg of the other. Horses may be trained with more or less success to adopt either gait ; but the distinction between natural pacers and natural trotters is a fairly sharp one (16). The physiological nature of the difference is quite obscure, but presumably it is of nervous origin.
39. From time to time mice are found hairless, with the skin thrown up into corrugated folds. Experimenting with such mice Mr Archibald Campbell found the condition to
* See also Davenport (107).
< 033 >
be a recessive, the presence of normal fur being a dominant. The fur grows at first normally and falls off as maturity is reached. Of 12 F2 mice 3 lost their hair. I am indebted to Mr Campbell for information respecting this interesting case, and for living specimens. The attempt to breed the recessives together failed, but in Gaskoin's case* naked parents produced young like themselves. From his account it appears that the young which he observed never grew their hair, but the fact is not absolutely certain from the description.
40. The normal condition and the “waltzing” habit in Japanese mice. The waltzers exhibit a peculiar vertiginous movement of the head when they come out into the light, and spin often with extreme rapidity, running after their tails till apparently exhausted.
Our knowledge of this case is derived from Von Guaita (135) and Darbishire (90). The dominance of the normal type is complete, and in F2 waltzers reappear. The F2 numbers obtained by Darbishire were 458 normals, 97 waltzers, where the expectation is 386 normals, 139 waltzers. The deficiency may perhaps indicate a complication, but more probably it is due to the greater delicacy of the abnormal mice, which was so great that all attempts to breed them together were unsuccessful.
41. Normal short hair and the long “Angora” hair Rabbit, Guinea-pig, and doubtless Cat (see Hurst, 157 ; Castle, 45 and 48 ; Sollas, unpublished ; Castle and Forbes, 55).
Castle (48), p. 64, gives important details as to the physiological nature of the distinction between the normal and “Angora” hair, which he regards as resulting from a special method of growth.
42. The rough or rosetted condition of the coat in the Guinea-pig dominates over the normally smooth condition (Castle, 48. ; Sollas, unpublished).
Castle found occasionally that animals partially rosetted occurred in F2.
* For references see Bateson, Materials for Study of Variation, 1894, p. 56. A good figure is given by Gaskoin, Proc. Zool. Soc. 1856.
< 034 >
43. Polydactylism occurred in a Guinea-pig, the offspring of normal parents, and ran an irregular course in its subsequent descent (Castle, 49).
44. The abbreviated tail of the Manx Cat is a dominant (more or less imperfect) to the normal tail (see Anthony, 2 ; Hind, 151 ; Davenport, 98 ; Kennel, 166, a).
45. Polydactylism is almost certainly dominant ; but, as in other types, irregularities doubtless occur.
For the study of heredity Fowls are especially well suited. In addition to their many colour-characteristics the various breeds present a great range and variety of structural features.
Among the long series of offspring which hens of the more fertile breeds produce, the descent of these characteristics can be watched in families of ample length. The chief papers dealing with Fowls are R.E.C. (19-22) ; Hurst (156) ; Davenport (101). The following is a list of the principal facts already elicited as to the behaviour of these structural features but much remains to be done.
46. Various shapes of comb, for example the rose comb and the pea comb, are both dominant to the single comb. The double or longitudinally split condition is also dominant to the unsplit.
See pp. 61-7. Many of the finer details in regard to the heredity of comb-shapes are not yet clear. The classification of the comb-types in the newly-hatched chickens is generally very easy, but in occasional strains forms intermediate between the pea and the single occur in F2, which may probably be due to subtraction-stages of the pea factor (q.v.). Some of the singles extracted in F2 from various crosses have lateral ”sprigs” - as fanciers say. It is not impossible that these irregular processes are due to additional minor factors, but they are subject to so much fluctuation that their descent would be very difficult to trace. The comb of the Silky fowl is a rose, + a trifid element which causes its posterior end to be divided into three irregular points. In F2 from Silky x Single, regular rose combs are produced in those individuals which have the rose factor without this trifid element.
Attention may be called to the dominance of the median splitting of the comb found in certain breeds, for the facts may have a bearing on the genetics of meristic characters. Splitting of the comb may occur in one of
< 035 >
several distinct ways. It may affect mainly the anterior portion, or the posterior. The split combs of established breeds have possessed ordinary dominance ; but a form of posterior splitting somewhat like that of the F2 from Breda x Single occurred apparently as a mutation among extracted singles, and exhibited a curious genetic behaviour suggesting irregularity of dominance (20, pp, 108 and 113).
47. The normally webbed feathers are dominant to the peculiar feathers of the Silky fowl.
48. Extra toe is usually dominant to the normal four-toed condition, but exceptions occur.
This irregularity of dominance is exhibited by all cases of polydactylism yet studied in birds or mammals. It seems to be a property of certain strains. Some families run a perfectly regular Mendelian course, others contain members with only the normal four toes, which are yet capable of transmitting the extra toe. The numbers in such families are not favourable to the suggestion that the irregularity is caused by a definite disturbing factor.
49- Crest is dominant to no crest.
F2 may contain individuals with crests far larger than those of the parent crested breed, a fact which suggests that in breeds with small crests (e.g. Silky) the full development of the crest is kept in check by some other factor.
50. Feathered leg partially dominates over clean leg.
Both Hurst (156) and Davenport found dominance very irregular. F1 is intermediate, and traces of leg-feathering are occasionally seen in the offspring of clean-legged birds.
51. “Frizzling,” or turning back of the feathers, is dominant to the plain straight feathers of the normal.
52. Normal size of feathers on the hocks, or tibio-tarsal region, is dominant to elongation of these feathers to form quills-the “Vulture-hock” of fanciers.
53. Muff, or tuft of feathers at sides of the bill and throat, as in Faverolles, is dominant to no muff, as in ordinary breeds.
54. Imperfect development of coccyx and tail-feathers with absence of tail, as in “Rumpless” fowls, is dominant to the normal development of those parts.
This interesting case was investigated by Davenport (101). It is exactly comparable with that of the Manx Cat.
< 036 >
55. Certain breeds (Houdan, Polish, Breda) have an extraordinary development of the nostril, which is patulous, with alae horizontal instead of curving downwards. This peculiarity is recessive to the normal (Davenport, 101). Hurst has observed the same thing and Mr Punnett and I have similar evidence from the Breda. Davenport states that in his experience the “high” nostril is never combined with a fully developed comb.
56. The tendency to go broody and sit on eggs dominates over the absence of this instinct, characteristic of several Mediterranean breeds. There is probably segregation in regard to these two dispositions, but this cannot yet be asserted positively.
In regard to fertility as measured by egg-production there is as yet no clear evidence.
57. The loud and penetrating shrieks which the cocks (and to a less degree the hens) of an Egyptian breed give out when caught, were reproduced almost exactly by the F1 generation from a cross with a non-shrieking breed. Though numerical data in regard to such a character are scarcely attainable, there is little doubt of the segregation as evidenced by F2.
58. The normal foot is dominant to the webbed condition of the toes which sometimes occurs as an abnormality (Staples-Browne, 254).
Mr J. L. Bonhote tells me that in his experiments webbed birds have produced normal offspring. He is making further experiments with this family.
59. The “shell,” or turning-back of the head-feathers of the Nun is dominant to the normal plain head (ibid.).
60. Birds with normal, 12-feathered tails crossed with the many-feathered Fantail give intermediate numbers in F1. In F2 12-feathered tails reappear, but, so far, no real Fan has come from the cross-breds. Mr Staples-Browne, to whom I am indebted for this information, will publish a complete account of his evidence. He tells me that the extracted 12-feathered birds do not breed true, but may throw birds with 13 or 14 feathers.
< 037 >
61. Crest is dominant to plain-head, as the non-crested condition is called by fanciers (R.E.C. 19, p. 131 ; Davenport, 105).
The type of crest which fanciers admire consists of feathers neatly laid down over the head. To produce such birds crested individuals are bred with plane-heady, and it is clear that the exhibition type of crest is a heterozygous form. When crested birds are bred together it is said that an ugly, standing crest frequently is produced, and presumably this is the homozygous type of crest. The mating of two crested parents is by several authors said to give rise to some bald birds. Other writers (e.g. Blakston) have ridiculed this statement and formerly I was inclined to regard it as a mere exaggeration, but Davenport in his recent paper mentions bald heads as sometimes occurring among his crested birds. He has kindly supplemented his published account with the statement that the bald patch is an area “on the back of the head varying from four to six millimetres in diameter practically without feathers and remaining featherless throughout life. The crest, however, on top of the skull remains perfectly evident, and often baldness can only be detected by blowing the feathers.” In no case was such a bald patch found in a plain-head.
The bald patch on the occiput is recognized by Blakston (Cage Birds, p. 104) as a property of crested birds, and presumably the “balds” alleged to come from the mating of two crests are birds homozygous for crest-factor, in which the crest stands up and allows the bald patch to be seen. Davenport had a crested bird without any bare patch, and he found that the feathering in this region was due to a separate dominant factor.
Animals and Plants in which Colour-Characters have been shown to have a Mendelian Inheritance.
The phenomena of colour-inheritance are complicated in several ways. Some of these complications which are of great importance and interest will be considered in subsequent chapters. It is, however, convenient to enumerate the genera in which Mendelian heredity has been observed in order to illustrate the scope of the principle. The following list of genera contains the chief of those in which heredity according to a Mendelian system has been shown to occur. In some of them as the result of extensive research many Mendelian features of colour have been discovered, and the existence of numerous colour-factors is demonstrated. In others only one such factor for colour has been detected.
< 038 >
Brassica (Turnips and Swedes).
Lathyrus (Sweet Pea).
Orchids (several genera).
Zea (Maize, Indian Corn).
Lepidoptera, various (Silkworm ; Abraxas grossulariata ; Angerona prunaria &c.).
Coleoptera (Lina ; Leptinotarsa ; Crioceris).
For the convenience of readers acquainted with the phenomena in outline and desirous of pursuing the subject further the following, brief annotations are placed here. Until the chemistry of pigmentation is better understood, a comparison between the behaviour and properties of the several types cannot be instituted with much confidence.
Antirrhinum. Wheldale (303) has shown that the lowest or hypostatic factor dominant to albino gives yellow in the “lips” of the flower ; the addition of various other factors produces anthocyan reds which superposed on the yellow give deep crimson red colour. A second series of reds, more purplish or magenta in tint (colour of wild A. majus), results from addition of a factor which in absence of anthocyans gives an ivory colour. This ivory is epistatic to yellow. It is remarkable that the lowest anthocyan factor gives red in the tube with a tinge in the lips, while the addition of the next above it gives the self-coloured flower.
There is also a white-tubed type of each colour-combination (“Delila” of de Vries, 298).
All the factors except those for yellow and ivory can be carried by the albino. Among the reds several heterozygous combinations can be recognized. The heredity of striping is still under investigation.
Atropa Belladonna. The normal dark-fruited type is dominant to yellow-fruited (de Vries, 290 ; Saunders, 19).
< 039 >
Brassica. White chromoplasts dominant to yellow in Swedes and Turnips (Sutton, 262).
Clarkia elegans. Common magenta-red dominant to salmon pink (Bateson and Punnett).
Coreopsis tinctoria. Ordinary yellow type dominant to var. brunnea with brown flowers (de Vries, 290). The brown flowers like those of Cheiranthus (Wall-flower) are no doubt due to presence of much dark anthocyan, and the case is probably one in which the development of little anthocyan dominates over the development of much anthocyan (cp. Lathyrus, Primula, &c.).
Datura. Purple in flower or stem dominant to white flower and green stem (de Vries, 290 ; Saunders, 19).
Gossypium (Cotton). Dominance of many colour-characters in plant, flower, and seed (Balls, 6). F2 details not yet published.
Helianthus. Purple disk dominant to yellow disk (Shall, 241).
Hordeum (Barley). Black pigment in paleae dominant to its absence (Tschermak, 270 ; Biffen, 30).
Hyoscyamus niger annuus x H. rager pallidus were found by Correns (69) to give F1 flowers of intermediate tint.
Lathyrus (Sweet Pea). Anthocyan colours dominant. Purples dominant to reds. Colour depends on two complementary factors. Yellow chromoplasts recessive to colourless. Facts fully described in later chapters. Plants with coloured flowers have dark seed-coats. Whites have seed-coats colourless.
Lychnis ; F1 between L. diurna and L. vespertina has flowers of intermediate tint ranging through many grades (see de Vries, 290 ; Correns, 69 ; R.E.C. 19). Segregation imperfectly studied.
Matthiola (Stocks). Colours as in Sweet Pea (R.E.C. 19-21 ; Tschermak, 278 ; Correns, 61). For colours of seeds see R.E.C. 19 .
Mirabilis. Colours consist of a complex series of reds and yellows, the interrelations of which are not yet clear (see Correns, 67, 74, 77) Miss Marryat's experiments (unpublished) prove the existence of a number of heterozygous forms.
Orchids. Dominance of anthocyan colour in Cypripedium is clear. In that genus it results from union of two complementary factors (Hurst, Gard. Chron. 1908, 1. p. 173). As regards distribution of colour the facts are complex, but several indications of Mendelian distribution have been recognized (Hurst, 153, 160). See p. 96.
Papaver. Presence of dark purple spot at base of petals dominant to the absence of such colour (de Vries, 290).
Phaseolus. The elaborate researches of Tschermak (271-3, 275, 278) have demonstrated the existence of numerous factors controlling the colour of the flowers and seed-coats in P. vulgaris, P. multiflorus and their hybrids. The flower-colours are purples, reds, and white, with a bicolour form of the red (“Painted Lady”). Colour has not yet been produced by union of
< 040 >
two whites, but on the analogy of the Sweet Pea such a result may be attainable. Otherwise the same rules apply generally. At least two sets of pigments take part in coloration of seed-coats : (1) brown, (2) purple. White-flowered plants have seed-coats unpigmented, and the bicolour flowers go with parti-coloured half-white seeds. The development of purple in the coats and the pattern in which it is deposited depend on various factors which can be carried by the albino. Various complications were met with (see originals). The cross between the two species showed some degree of sterility.
Similar results were obtained by Emerson (120, 121 ). The seed-coats of heterozygous plants in some cases were distinguishable. He also found green pods dominant to yellow pods (cp. Pisum).
Further facts, with scheme elucidating some of the curious ratios which the seed-colours may exhibit (e.g. 18 : 18 : 6 : 6 : 16) in F2, are given by Shull (242).
Phyteuma Halleri (dark violet) x P. spicatum (white) gave two types in F1, 5 plants being bright blue with violet tinge, 4 violet (Correns, 70).
Pisum (Edible Peas). Flower-colours of three types. (1) Purple. Standard a pale puplish white ; wings deep chocolate purple. (2) Pink. Standards pinkish white ; wings a fine salmon-pink. (3) White. F2 containing all three types is the usual 9 : 3 : 4 in order named. Mark in axils of leaves, if present, is purple in (1), red in (2), absent in whites. Tschermak experimented with a purple strain without the axil-mark, and found that, as in Sweet Peas, the factor for that character can be carried by the albino.