58 Bulletin American Museum of Natural History. [\o\. XXIII, 



and have been maintained with more or less modification by zoologists 

 down to the present time.^ 



In order to make his explanation of polymorphism conform with the 

 remainder of his hypothesis of heredity, Weismann is compelled to assume, 

 not only a set of germ-plasmic determinants for each of the sexes in ordinary 

 sexual dimorphism, but a double set for the female in the social Hymenoptera 

 and a double set for each sex in the termites. At least four sets must also 

 be postulated for ants with differentiated worker and soldier castes. This 

 is evident from the following quotation : ^ 



"So when instead of dimorphism, there is polymorphism, when, for 

 instance, the females of a species are similarly distinguished among them- 

 selves, and occur in two forms, this results, according to my idea, from the 

 double determinants becoming triple determinants. If there were workers 

 among the butterflies, and if these showed red color on the part of the wing 

 that is blue in the male and brown in the female, there would always be three 

 representative determinants present at a definite part of the extremely 

 elaborate and highly complicated germ-plasm; but only one of these would 

 become active during the development of the egg and sperm-cell concerned, 

 and would produce the patch of brown or blue or red scales in the wing. 



"According to this theoretical representation, every part of the body 

 that is differently formed in the males, females and workers is represented 

 in the germ-plasm by three corresponding determinants but in the develop- 

 ment of the eggs, never more than one of these attains to value — i. e., gives 

 rise to the part of the body that is represented — and the others remain 

 inactive. 



"Thus, then, the metamorphosis of the body parts of the workers of ants 

 and bees will have to be considered in connection with the fact that the males 

 and females whose germ-plasm contains favorable variations of the deter- 

 minants of the workers have a better prospect for the maintenance of their 

 successors than others which showed less favorable variations of such 

 determinants. The process of selection is the same as if the matter at issue 

 were the attainment of favorable adaptations in the body of the sexual forms ; 

 for in both cases it is, as I have once before said, not really the body that is 



1 The papers called forth by this controversy are here cited in the order of their publication: 

 Spencer: 'The Inadequacy of "Natural Selection",' Contemp. Rev., Feb. and March, 1893, 

 pp 153-166, 439-456; 'Professor Weismann's Theories,' ibid. May, 1893, pp. 743-760; Weis- 

 mann: 'The All Sufficiency of Natural Selection,' ibid., Sept., 1893, pp 309-338; Spencer: 

 *A Rejoinder to Professor Weismann,' ibid., Dec. 1893, pp. 893-912; 'Weismannism Once 

 More,' ibid., Oct. 1894, pp. 592-608. Weismann has also repeated or elaborated his views on 

 the polymorphism of the social insects in the following works: ' Das Keimplasma. Eine Theorie 

 der Veferbung.' Jena, Gustav Fischer, 1892, pp. 494-498; 'The Effect of External Influences 

 upon Development,' Romanes Lecture, London, Henry Frowde, 1894, pp. 29-48; and ' Vortritge 

 uber Descendenz-theorie,' Jena, Gustav Fischer. 2 Bd. pp. 101-118. For an aftermath of the 

 Spencer- Weismann controversy see also the discussion of the subject in ' Natural Science,' IV, 

 1894, by Ball (Neuter Insects and Lamarckism, pp. 91-97), and Cunningham (Neuter Insects 

 and Darwinism, pp. 281-289). 



2 The All-Sufficiency, etc., loc. cit., pp. 326, 327. 



