white; but to the statement that the hybrids have black eyes there was no exception in the cases, which exceeded three hundred, that I have bred. The hybrids never exhibit waltzing movements, and from the description of their coloration it is evident that they are never albinos. When such hybrids are bred together they produce a population which, considered from the point of view of its colour, falls into three categories: the first is black eye and coloured coat, under which heading come half of the population-mice, therefore, which resemble their parents; the second is pink eye and coloured coat, and includes a quarter of the population, individuals presenting the same features of eye and coat colour as those exhibited by the Japanese waltzing mouse; and the third, into which the remaining quarter falls, is pink eye and uncoloured coat-that is to say, it is albino. If we examine the offspring of hybrids from the point of view of their progression, we find that rather less than a quarter waltz, while the rest are normal. The waltzing habit in this population is not always associated with the same arrangement of eye and coat-colour as that with which it is associated in the pure Japanese waltzer, but may be presented by mice falling into any of the three colour categories. To return to the colour question, the albinos, which, it will be remembered, form a quarter of the population produced by mating the hybrids, breed absolutely true; the pink-eyed mice with coloured coats breed nearly true, and the black-eyed individuals with coloured coats produce, when paired together, albinos, pink-eyed mice with coloured coats, and black-eyed mice with coloured coats, but in what proportions I have not yet determined. Some of the facts which have come to light seem confirmatory of the Mendelian interpretation of these phenomena, while others are describable in terms of either Galton's or Pearson's formula of ancestral inheritance. I do not think, therefore, that I am justified in forming an opinion on the question of the relative validity of these two interpretations of the facts already observed, and until more data have been collected I do not propose to do so.

3. Experiments on Heredity in Rabbits. By C. C. HURST, F.L.S.

An inbred pair of albino Angoras was crossed reciprocally with an inbred pair of Belgian hares (F,), and the hybrid progeny were bred with one another for two generations (F, and F). Four characters were under observation, each of which was inherited independently of the other.

1. Angora Coat.-In F, the angora coat was always recessive to the normal coat, which was completely dominant. In F, and F, this pair of structural characters followed the Mendelian laws of segregation and gametic purity simply and without exception.

2. Albinism.-In F, the albino character was always recessive to the normal character, which was dominant, and in F, and F, followed the ordinary Mendelian rules.

2

3. Coat Colour.-In F, brown albino gave all with wild grey coats. In F, the hybrid greys bred together gave a ratio of 9 grey: 3 black: 4 albino. Experiments in F, proved that the black factor was not introduced by the original brown parent, but by the albino, which, though gametically pure as regards simple albinism, was at the same time carrying the distinct factor for black coat colour.

These results in rabbits confirm the important results already gained by

Cuénot in mice.

The F2 greys proved to be of four kinds-viz., pure grey, grey containing black, grey containing albino, and grey containing black and albino. The F blacks were of two kinds-viz., pure black and black containing albino. The F albinos were of three kinds-viz., albino containing grey, albino containing black, and albino containing grey and black. These results are in accordance with the Mendelian expectation, which is—

1G: 2G(B): 2G(A): 4G(B)(A): 1B: 2B(A): 1A(G): 2A(G)(B): 1A(B)

One point, however, remains to be cleared up, and that is, the absence of browns in F, and F,, these being in all cases apparently displaced by greys.

2

4. Dutch Markings. In F, the offspring from one albino were, as 8 rule, uniform in colour, like the coloured parent, while those from the other albino were all more or less marked with white on the fore extremities. In F2 and F, the uniform hybrids were, as a rule, constant, while the marked ones produced a proportion of true Dutch-marked rabbits, as well as the ordinarymarked and some uniform ones.

These results suggest that one albino contained the factor for Dutch markings while the other albino did not. Experiments are now in progress to work out this interesting question.

4. Experiments on Heredity in Fowls. By R. C. PUNNETT.

5. An Intermediate' Hybrid in Wheat. By R. H. BIFFEN.

In the majority of cases investigated up to the present one character of a given pair (the dominant) masks the other (recessive) in the first generation (F,).

In the case of Triticum Polonicum (Polish wheat) x T. turgidum (Rivet wheat) and its reciprocal, the hybrid does not show this sharp separation of dominant and recessive characters, it being intermediate between the parents with regard to certain pairs. Thus in Polish wheat the glumes, grain, and internodes of the spike are long, and it ripens early; in Rivet wheat the glumes, grain, and internodes are short and it ripens late; whilst the hybrid has an intermediate length of glumes, grains, and internodes, and also ripening period. The hybrid has therefore a distinct character of its own in which one cannot recognise definite dominant and recessive characters. However, in the following generation (F) the splitting is of the type which Mendel's work has made so familiar to us. We find long, intermediate, and short glumes, early, intermediate, and late ripening, &c., in the proportion of 1: 2: 1, showing that, in spite of the fact that there is no marked distinction into the usual dominant and recessive characters, the gametes are still pure with respect to the characters they carry.

6. Experiments on the Behaviour of Differentiating Colour-characters in Maize. By R. H. LOCK, B.A.

Maize as grown in Ceylon is of a flint variety, but shows a complex mixture of the colours white, yellow, blue, and red.

White, yellow, and blue grains occur mixed in the same cob. The red colour appears either in all the grains of a cob or not at all.

The problem attempted was to discover, by the method of growing the offspring obtained by definite pollination, how far these several colour-characters conform to Mendel's law.

When a few yellow grains appear in an otherwise white cob such grains must be supposed to be the result of accidental impregnation of pollen bearing the yellow character. On sowing these grains and fertilising the female flowers with the pollen of a white variety the plants yield 50 per cent. of yellow and 50 per cent. of white grains within the limits of error due to the size of the samples afforded by the cobs, which contain as a rule from 300 to 800 grains.

Blue grains, appearing in the same way in an otherwise white cob, when treated as above yielded, as a rule, a much smaller percentage of blue grains-e.g., 30 per cent.-and the proportion is also more variable. The intensity of the blue colour also varies considerably. But on sowing the grains from such a cob, i.e., one with 30 per cent. of blue grains, and pollinating with white, it was found that the blue colour reappeared in the offspring of approximately 50 per cent. of the original grains; i.e., in those of the 30 blue grains, and in those of 20 out of

the remaining 70 white grains. It appears, therefore, that Mendel's law (as enunciated by Correns, namely, that the germ cells represent in equal numbers all possible combination of paired characters, no two members of the same pair occurring in the same germ cell) is truly followed, but that there is considerable irregularity in dominance.

The red colour does not appear as an immediate consequence of cross-fertilisation, and is, therefore, more difficult to deal with. It appears only in the offspring of the cross (white x red), being situated in the pericarp, and, therefore, a plantand not an endosperm-character.

6

The evidence shows that if a 'red-coated' plant of unknown parentage is crossed with white and the red-coated' offspring again crossed by white, 50 per cent. of the plants resulting from this cross will show the red character and 50 per cent. not. It is therefore highly probable that Mendel's law is

followed in this case also.

Specimens were exhibited showing various regular combinations of the characters dealt with.

7. Experiments on Heredity and Sex-determination in Abraxas grossulariata. By Rev. G. H. RAYNOR and L. Doncaster.

In the currant moth (Abraxas grossulariata) a rare variety occurs known as var. lacticolor or flavofasciata, found hitherto in the female sex alone. The Rev. G. H. Raynor has bred this species for several years, and the following is a summary of his experiments.

Var. lacticolor is recessive in the Mendelian sense, not appearing at all in the first cross. In the offspring, of heterozygotes paired together half the females are lacticolor, the remainder of the females and all the males being normal grossulariata (1). For example, the numbers bred in one family of this class were 25 normal, 14 normal 9,9 lacticolor ; in another, 22 normal 8, 9 normal, 9 lacticolor 9.

When, however, a lacticolor is paired with a first cross (namely, Lo̟ × G(L) ♬ ), among the offspring, not only some of the females, but also some of the males are lacticolor (2). The numbers available are not yet enough to determine with certainty what are the proportions; in one family there were 10 normal, 6 lacticolor 3, 4 normal and 2 lacticolor .

The facts may be summarised in genealogical tables thus:

The experiments are of importance in relation to Castle's hypothesis that gametes bear one or the other sex, and that certain somatic characters may be coupled with a given sex in the gametes. The hypothesis, if somewhat modified, is in excellent accord with the facts; but until we know the result of the pairing lacticolor cross-bred it would be premature to draw far-reaching conclusions.

(In this exhibit were also included, as being of interest to entomologists in general, some of the most striking aberrations Mr. Raynor has reared of this extremely variable species.)

8. Experiments on Heredity in Web-footed Pigeon. By R. STAPLESBROWNE.

Cross made.-Web-footed pigeon x Nun q.

Characters.-1. Foot character.-The membrane between the digits is extended as far as the toe-nails. This character appears suddenly in a strain of pigeons, birds possessing it being bred from parents with perfectly normal feet.

2. Head character. Shell.' A tuft of reversed feathers at the back of the head, forming the 'shell.' This character has been bred true in the 'Nun' pigeon for many generations.

The web-footed character appears on hatching, the 'shell' on feathering. The experiments were started in 1902, and include birds hatched up to July 16, 1904.

9. Fowls and Sweet Peas. By W. BATESON, F.R.S.

10. Report on the Occupation of a Table at the Zoological Station, Naples. See Reports, p. 300.

11. Report on the Index Animalium.'-See Reports, p. 297.

12. Report on the Influence of Salt and other Solutions on the Development of the Frog.-See Reports, p. 288.

13. Report on the Colour Physiology of Higher Crustacea.
See Reports, p. 299.

14. Report on the Coral Reefs of the Indian Ocean.-See Reports, p. 298.

15. Report on the Occupation of a Table at the Marine Laboratory, Plymouth.-See Reports, p. 297.

16. Report on the Zoology of the Sandwich Islands.-See Reports, p. 298.

17. Report on the Madreporaria of the Bermuda Islands.
See Reports, p. 299.

MONDAY, AUGUST 22.

The following Papers were read :—

1. Egyptian Eocene Vertebrates and their Relationships, particularly with regard to the Geographical Distribution of Allied Forms. By Dr. C. W. ANDREWS.

2. Normentafeln' of the Development of Vertebrata.

By Professor F. KEIBEL.

The Normentafeln ' are meant to form a sound foundation for the comparative anatomy of the vertebrata.

Each Normentafel gives (i.) an almost complete series of drawings of embryos; (ii.) tables concerning the degree of development of the various organs, with a few illustrations in the text; the tables furnish data concerning the exact time of appearance of organs which permit of criticism of 'biogenetical law'; (iii.) a bibliography arranged alphabetically and according to subject.

Those already published relate to the pig, chicken, ceratodus, and lizard. The frog by Dr. Kopsch, and the rabbit by Professor Minot and Dr. Taylor, will shortly appear.

3. On the Embryos of Apes. By Professor F. KEIBEL.

The embryos of the apes, of which I have the pleasure of showing you the drawings and photographs, belong partly to the collection of Professor Selenka, who unfortunately died before having completed his investigations. I am indebted to Professor Hubrecht for the rest. Specimens are shown of orang-utang, gibbon, macacus, and semnopithecus. The youngest embryo in my possession measured 13 mm., and resembled in its development a human embryo of about 12-14 days, the oldest measuring 3 cm.-corresponding to a human embryo of 10-12 weeks. The result of this examination, which proves of the greatest general interest, is the close resemblance which the ape embryos show to the human embryos in relative stages. This result Selenka concisely emphasised shortly before his death in the Biologische Centralblatt.'

Various differences-apart from the tail-are to be traced by closer study, not only between human and simian embryos, but also between the different species of apes. I do not doubt that it will be possible in course of time to detect differences in the early stages of these various embryos. We are already able to do this without difficulty from the fourth to the fifth week.

4. On Professor Loos's recent Researches on Ankylostoma (the Miner's Worm). By A. E. SHIPLEY, F.R.S.

Professor Loos, of Cairo, through Dr. Elliot Smith, communicated certain of his recent observations on Ankylostoma duodenale.

Particular stress was laid upon the fact that the larvæ can penetrate the unbroken skin without causing any visible lesion of the part.

After thus passing through the skin the larvæ enter the lymph vessels and the