36 



NATURE 



[January 12, 1922 



we may look for much light from them. The con- 

 viction seems to be growing that at all events a slow 

 and gradual change in the germ-plasm may be pro- 

 duced by altered conditions, although the main lines 

 of heredity may be determined by ancestral nature. 



Many obscure facts have to be taken into account, 

 and it is no wonder that the author of the book 

 before us finds it an easy task to show how unsatis- 

 factory are the various theories of heredity and 

 variation that have been suggested. He therefore 

 puts forward a new one, which appears to have 

 taken its present form on account of the renewed 

 attention attracted to the effect of the chemical pro- 

 ducts of one organ on the activity of another — on 

 account of the discovery of the mechanism of pan- 

 creatic secretion by Starling and the present re- 

 viewer in 1902. The hypothesis that every tissue 

 of the body gives off its own specific chemical pro- 

 duct, and that this product has its effect to a greater 

 or less degree on all other tissues, was put forward 

 by Brown-Sequard and d'Arsonval as early as 1869, 

 but was thrown somewhat into discredit by the un- 

 critical use of it by the former. Dr. Cunningham 

 gives the credit of the first suggestion to Claude 

 Bernard in 1855, but we regret to have been unable 

 to find the statement referred to. It would be of 

 much interest to have the exact reference. 



The hormone theory of heredity may best be given 

 in the words of its propounder : ' ' We have within 

 the gonads numerous gametocytes whose chromo- 

 somes contain factors corresponding to the different 

 parts of the soma, and these factors or determinants 

 may be stimulated by waste products circulating in 

 the blood and derived from the parts of the soma 

 corresponding to them" (that is, to the determin- 

 ants). Thus the effect of chemical products on any 

 particular organ or tissue in the soma is to be sup- 

 posed to be exercised in the same way on the " de- 

 terminant " in the germ-cell which afterwards gives 

 rise to such organ in the progeny. For example, the 

 exostosis on the frontal bone of stags, formed as a 

 result of repeated butting, would give off products, 

 not necessarily different in kind from those of bone 

 in general, but in increased amount, and thus 

 stimulate the corresponding factors in the germ-cells. 

 While it would perhaps be rash to deny the possi- 

 bility of a process of this kind, the objection might 

 naturally be made that when applied to the in- 

 heritance of bodily structures in general, or of 

 changes in them, it argues so enormous a variety 

 of " hormones " as to seem almost incredible. Not 

 only so, but the chemical product of each organ and 

 tissue must act on the germ-plasm in a way which 

 leads to the formation of a tissue like that by which 

 the hormone was formed. In view of the differ- 

 ence between the structure and activity of the germ- 

 NO. 2724, VOL. I09J 



plasm and those of the various constituents of the 

 soma, are we justified in supposing that a particular 

 chemical compound will affect both in the same way 

 or even in a similar way ? Moreover, waste products 

 would be expected rather to have a retarding than 

 a stimulating influence on similar reactions. But it 

 might be held that the hormones in question are not 

 waste products in the ordinary sense, and that the 

 precise name is immaterial. 



The possibility cannot be denied that, however 

 inaccessible to nervous action the germ-plasm may 

 be, it must be affected by chemical agents in the 

 blood. Indeed, Stockard's experiments, to mention 

 a single instance only, show that this is so with 

 alcohol. But in such cases the effect is of a more 

 or less generalised nature on the progeny, and 

 the existence of tissue products of the kind 

 demanded by Dr. Cunningham's theory is not yet 

 demonstrated. As Prof. Swale Vincent has pointed 

 out, it is remarkable how few ' ' internal secretions ' ' 

 have actually been shown to exist as chemical in- 

 dividuals. Although it may savour too much of 

 mysticism, it is open to question whether the original 

 form of Dr. Cunningham's theory, in which " in- 

 fluences " were spoken of, might not be the more 

 cautious and wiser one as yet. The loss of differ- 

 entiation in the growth of tissues in vitro, except 

 in the proximity of another tissue, no doubt indi- 

 cates some kind of influence by one tissue on 

 another. But the fact that this influence disappears 

 when the new cells wander away into the culture 

 fluid is difficult to reconcile with a chemical product. 

 Notwithstanding this objection, the theory must 

 be given the credit of introducing functional or 

 physiological considerations into the problem, as 

 does the somewhat similar one of Delage. Most 

 theories seem to be content with the purely struc- 

 tural view of rows of determinants in chromosomes 

 and the shifting about of these. The difficulty in 

 those theories which limit the transmission of heredi- 

 tary characters to the chromosomes is that these 

 exist as distinct entities only at the time of karyo- 

 kinesis, while even the nucleus itself is but a part 

 of the cell in functional relation with the whole. 

 Whether the chromosome view is necessarily in- 

 volved in Mendel ian interpretations is subject to 

 doubt, however significant the experimental facts 

 may seem. Bateson appears to be unconvinced, and 

 states that the results of such experiments have not 

 solved the problem of adaptation, while Brachet 

 has obtained evidence that the ovum, when fertilised 

 after removal of the nucleus, can transmit characters 

 of the female parent. 



Dr. Cunningham appears to be justified in his 

 complaint that, although his theory was published 

 in 1908, later writers have put forward similar 



