284 C. O. WHITMAN. 



' xVnnals and Mag. Nat. Hist./ April, 1873 The feature 



in whicli they differ from the nuch-i of cleavage-segraeiits is this, 

 tliat no area becomes segmented around them." These nuclei 

 multiplied not hy division of pre-existing nuclei, but by inde- 

 pendent segregration. Towards the close of the epiboly they 

 were very numerous and scattered over the entire surface of the 

 egg. They were found to form " a large portion of the deeper 

 substance of the embryo.'" 



In another place {\^) Lankester expresses the opinion that 

 the cells of the perimorula (Gammarus fluviatilis) arise as 

 "isolated cells,^' in the same manner os the " autoplasts " in 

 the Cephalo])ods. Kowalevsky (ttoit) reports similar forma- 

 tions found in the yolk under the edge of the blastodisc, in the 

 egg of Pyrosoma. Their origin was unknowi 



The phenomenon of " free nuclei " has long been known 

 among Arthropods. The opinion formerly entertained (Weis- 

 mann, iw) that these nuclei were "new formations,^' having 

 no genetic relation with pre-existing nuclei, has not been corro- 

 borated by the later and more trustworthy investigations on this 

 point. 



Metschnikoff (Ufjttt) and Mayer (118) have traced the 

 nuclei of the blastoderm directly to the primary egg-nucleus 

 in Aphis, Cecidomyia, Miaster, &c. Besides, Ganiu (50), van 

 Beneden and Bessels (14), Hiickel (65, Peneus), Mayer (113, 

 Eupagurus), and others have shown that in many Crustacea a 

 genuine cleavage takes place. Ludwig (112) has established the 

 same fact in reference to spiders. The studies of Kowalevsky 

 (85), taken in connection with the supplementary observations 

 of l)ohrji ifyi), show clearly that the larger part of the yolk in 

 the eggs of many insects is subject to cleavage. Ajiis mellifica 

 furnishes a good illustration. Kowalevsky's statements is (-ll-) as 

 follows : — One finds on sections clearly outlined nuclei, like those 

 seen in the cells of the blastoderm. They are found in all parts 

 of the yolk, but are most numerous near the surface, just under 

 the blastoderm. The maximum number of the nuclei, which at 

 tii'st are few, is reached about the time of exclusion. These 

 nuclei were supposed to disappear with the yolk, taking no part 

 in the embryonic tissues, and consequently having no other 

 ])hysiological function than that of hastening the dissolution of 

 the yolk. Like the nuclei of the blastoderm, they were sup- 

 posed to be derived from the egg-nucleus. In the case of 

 Le])idoptcra also (^;' ) the entire yolk, beginning at the periphery 

 and progressing towards the centre, breaks up into " Dotter- 

 ballen," in each of which a clear spot is seen. Dohrn (tiir) 

 goes farther, recognising, the " clear spot " as a nucleus and 

 the " Dottt-rballen " as cells. In coiulusiun Dohrn remarks, 

 p. 122 : " ri-oiU all tills it ijs bertaiu that the yolk in the egg of 



