242 



SCIENCE 



N. S. Vol. XXX. No. 764 



Many external and internal conditions which 

 do not appear to affect the transmission of 

 characters are able to influence the expression 

 of characters. Differences of heat and light, 

 as well as of foods, chemicals and internal 

 secretions (enzymes) are now known to induce 

 changes in the expression of characters. 

 "Without the thyroid gland the remainder of 

 the body does not complete its growth. Cas- 

 trated animals may grow to abnormal size, but 

 fail to develop secondary sexual characters. 

 Parasitic fungi, insects and mites induce the 

 development of galls and other changes in the 

 habits of growth of their host plants. Horti- 

 cultui-ists have learned that scions have defi- 

 nite influences upon the characters of the roots 

 on which they are grafted. Plants grown 

 under new and unaccustomed conditions often 

 show wide ranges of individual variation, re- 

 calling many ancestral characters to expres- 

 sion. There is no ground for denying the 

 possibility that similar reversions might follow 

 impregnations by diverse types. 



The fact that Ewart found only faint 

 stripes on colts that followed his zebra hy- 

 brids does not destroy the older evidence that 

 more vivid stripes, as well as peculiarities of 

 mane and hoofs, followed the earlier instances 

 of hybridization with the quagga, an animal 

 of a different species, now reckoned as extinct. 

 Atavistic changes have been found to be more 

 frequent and more pronounced in some series 

 of hybrids than in others, even in the same 

 group of organisms. When our United States 

 Upland cottons are crossed with the Kekchi 

 cotton of Guatemala the hybrids have lint 

 shorter than either parent, but when the 

 Kekchi cotton is crossed with the Egyptian 

 the lint is as long as in the Egyptian parent, 

 or even longer. With hybrids, as with the 

 parent stocks, external conditions often ap- 

 pear to have a definite influence upon the ex- 

 pression of characters. Hybrids of the same 

 parentage may show Egyptian habits of 

 growth under some conditions and Kekchi 

 habits under others.' 



' " Suppressed and Intensified Characters in 

 Cotton Hybrids," Bulletin 147, Bureau of Plant 

 Industry, U. S. Department of Agriculture. 



The frequent occurrence of stripes in mules 

 may also mean that hybridization has a tend- 

 ency to induce this form of reversion. 

 There is no reason to suppose that the stripes 

 come from the asinine ancestry alone. 

 Striped mules seem to be especially common 

 in tropical countries, where unfavorable cli- 

 matic conditions and the mixing of the breeds 

 of horses may further increase the tendency 

 to reversion. A vividly striped mule of a 

 yellowish, zebra-like ground-color, seen at 

 Cordoba, Mexico, a few years ago, gave me a 

 better appreciation of the extent to which such 

 reversions may sometimes be carried. The 

 conspicuous markings of the legs happened 

 to be seen first, the more uniform body of the 

 animal being concealed by a wagon. I stepped 

 into the street expecting to find a traveling 

 managerie. 



The possibility of bringing telegony into re- 

 lation with other induced forms of reversion 

 adds nothing, of course, to the evidence that 

 reversions are induced by previous hybridiza- 

 tion. Facts may establish theories, but the- 

 ories do not establish facts, except as they 

 lead to further observation. If enough rever- 

 sions of the same kinds are found to occur 

 without previous hybridization it will be rea- 

 sonable to view all the alleged eases of teleg- 

 ony as coincidences. If reversions prove to 

 be more frequent after hybridization telegony 

 wiU be established, though its manifestations 

 may not be otherwise different from rever- 

 sions that occur without hybridization. 



Negative evidence may show that induced 

 reversions are rare, but does not affect the 

 authenticity of particular cases, as Ewart 

 himself perceived. Xenia also has appeared 

 to be very rare in nature at large, but its fre- 

 quent occurrence in maize is no longer doubted, 

 and an explanation has been found in double 

 fertilization, enabling the endosperm as well 

 as the embryo to share the characters of the 

 male parent. We need not feel obliged to dis- 

 credit facts like those collected by Darwin and 

 other students of heredity merely because an 

 erroneous theory suggested the name telegony. 



Pearson's plan of proving or disproving 

 telegony by a statistical study of the degrees 



