[12] Biol. Bull., XX, 1910, p. 67.
[13] As to the interrelations of these three forms, Tower states (1906, p. 18) that angustovittata, which he reared from undecimlineata, is intermediate between it and signaticollis. Compare Stål, "Monogr. des Chrysomélides," 1862, p. 163; and Jacoby, Biol. Centr. Amer. Celeopt., vi, Pt. 1, p. 234, Pl. xiii, fig. 20; Tab. 41, fig. 15; ibid., Suppl., p. 253. All these forms are evidently very closely related, and the delimitation of species is quite arbitrary. Jacoby indeed suggests that undecimlineata may be a variety of decemlineata.
[14] Gortner, Amer. Nat., Dec., 1911, XLV, p. 743.
[15] Mutations, Variations, and Relationships of the Oenotheras, Carnegie Institution Publication No. 81, 1907, pp. 61-64.
[16] Macdougal, D. T., "Alterations in Heredity induced by Ovarial Treatments", Bot. Gaz., vol. 51, 1911, p. 241.
[17] Payne, Fernandus, Biol. Bull., XVIII, 1910, p. 188, and ibid., XXI, 1911, p. 297.
[18] See especially, Mutation et Traumatismes, Paris, Felix Alcan, 1908.
[19] J. H. Powers, "Morphological Variation and its Causes in Amblystoma tigrinum." Studies from the Zoological Laboratory. The University of Nebraska, No. 71, 1907.
[20] In connexion with this case I would refer the reader to some remarkable observations of Dr. T. A. Chapman on various types of larvae which he reared from the moth Arctia caja (Ent. Rec., IV, 1893, p. 265, and following parts). From a single mother he raised a great diversity of forms, some which fed up rapidly and passed through their development without assuming certain stages, and others which were, as he called them, "laggards," moulting more times than their brethren and developing at a much slower rate. It is greatly to be hoped that such a case may be critically investigated by analytical breeding.