JfEMOIKS OF TUE NATIONAL ACADEMY OF SCIENCES. 425 



Iiilar medium. This view does uot commend itnelf to me. It is opposed to my own observations on similar iinilci 

 in the Spiders. It does not fit in with onr knowledjje of the nature of the ovum, and can not bo reconciled with tlio 

 8e};mentation of hucIi types ax Spiders, or even Gupaj^urua, with which tlio segmentation in insects is uudoubt«dly 

 closely related. — {Comp. Embryology, Vol.i.p. ll'.l. ) 



This discussion seems to have arisen from a confusion of tlie morj)liolo^ical and pliysiolo};ical 

 significance of tlie cell. The segmentation of the nucleus and it8SuiT<)uii(liiigi)rotoi)lasm is |)lainl,v 

 the only important phenomenon, and the segmentatior of the yolk is not merely a secondary process, 

 but in many cases a wholly unnecessary one, as we see in the early phases of many Decapods. 

 In these cases the individuality of the yolk i)yramid is temporarily sacrificed or suhordiiiatcd to 

 that of the true cell, which is surrounded by unsegmented yolk. Later, when the yolk has lu'comc 

 divided, the yolk segment or pyramid is gradually reduced until we get the superficial, embryonic 

 cell, with more or less definite boundaries. All the elements of the egg, whether suiierhciid or 

 aino'boid, are clearlj- to be regarded as cells in a fundamental, physiological sense, as shown by 

 the part which they play in the development of the embryo. 



SECONDARY SEGMENTATION OF THE YOLK. 



There are traces of a secondary segmentation of the yolk in Alpheus during the second, third, 

 and fourth stages; that is, from the period of invagination to the outlining of the primary append- 

 ages. The yolk spheres arrange themselves in spherical clusters or balls, so characteristit; of the 

 early development of nearly all the Arthropods. The yolk ball contains at least one yolk nucleus 

 with ]>erinuclear protoplasm and corresponds to a yolk pyramid, being a cell in the same sense as 

 the latter. Various jjhascs of this secondary segmentation may be seen by glancing over Pis. xxx- 

 XXXV In one egg, which I sectioned just prior to invagination (Fig. 46), there appears a segmen- 

 tation of the yolk around the central nuclei. 



Kobret/ky attributes a morphological value to the secondary segmentation of the yolk in 

 Arthroi)ods, supposing it to be connected with the spreading and final establishment of the ento- 

 blast. The secondary yolk pyramids or giant endoderm cells, which form the lining of the midgut 

 of the embryo crayfish, he compared with the Dotterhallen of Oniscns and I'alaMiion. In PalaMnon 

 the food yolk breaks up into round or polygonal pieces soon after the blastoderm is formeil, while 

 in Oniscns certain cells pass into the yolk from the keel or germinal eminence and gorge them- 

 selves with the yolk substance until they form large balls, which re|)resent the endoderm (Darm- 

 driisenzellen). It is stated, however, by Nusbaum (4'1) that a part of the endoderm of Oniscns 

 which gives rise to the gastric gland arises from i)rimitive inesoblast, and in insects the endoderm 

 is formed independently of the yolk cells. The history of the yolk cells and of the wandering cells 

 will be discussed in another section. 



VI — CELL DEGENERATION. 



The rapid and often extensive breaking up and final disappearance of embryonic cells in the 

 course of Arthropod development is a very remarkable phenomenon, and strange to say, it has 

 almost escaped attention up to the present time among the Decapod Crustacea. A study of this 

 subject in Alpheus, Astacus, and Ilomarus has convinced me that the i)eculiar bodies described 

 as secondary mesoderm cells in the crayfish (.")4) correspond to the degenerating, sporelike par- 

 ticles which characterize similar stages in the development of both Alpheus and Ilomarus. 



In some early notes on the development of Alpheus I called these nuclear fragments "sjjores" 

 (22), but the term is inappropriate if we are dealing with cells in the process of dissolution, as is 

 undoubtedly the case. The anomalous "secondary cells," which have been a sort of outstan<1ing 

 puzzle toembryologists, receive, in my opinion, a more reasonable explanation on the ground that 

 they represent degenerating elements. This view is supported by a comparative study of embry- 

 onic growth in other Arthropods. 



Alpheus. — Degenerating cells are present in Alpheus saulcyi in considerable nund)ers wlicn the 

 nauplius apjieiidages are budding and increa.se for a short time beyon<l this iteriod. They continue 

 in greater or less cpiantity until six to eight pairs of postoral apjx'ndages are formed, when they 

 disappear from the embryo almost (completely. They vary in size from small refringent particles 

 to spherical nnvsses as large as ordinary nuclei, or even larger. Many nuclei, instead of having 

 the normal appearance, in which the chromatin has the form of a coil or a reticulum with stellate 



