[103] EMBRYOGKAPHY OF OSSEOUS FISHES. 557 



closely euougli investigated. lu fishes, these caryokiuetic or nuclear 

 ligiues are found by lue in the segmenting blastodermic cells at a late 

 stage, or after the blastoderm has already covered half of the yelk 

 sphere. Similar facts are recorded of the germinal area of the blasto- 

 derm of higher ty[)es by Rauber. 



Hardly any one would attenii^t to dispute the ground taken by liau- 

 ber that cell-division is the result of growth, addition, or further dif- 

 ferentiation of plasma, and not the reverse, else we should very soon be 

 brought to the absurd position of assuming tliat cell-division might go 

 on indetinitely without the addition of new matter, which we know is 

 not the case. In the preceding section on nuclear displacement, which 

 is very commonly manifested, so commonly, indeed, that perhaps rel- 

 atively very few perfectly holoblastic or evenly segmenting ova are 

 known, we have an extensive piece of evidence in favor of the doctrine 

 that growth must precede segmentation. Just exactly how the passive 

 deutoplasma is broken down and raised to the grade of protoplasm we 

 do not know in detail in many cases, but we do know enough of the 

 ])rocess in others at least to infer what may be its general type or mode 

 in all. In general terms, it may be stated that the deutoplasm or yelk is 

 more or less closely invested or inclosed by the germinal protoplasm, 

 the function of which is clearly appropriative either by direct contact 

 or by the agency of fluid ferments acting by means of what Foster would 

 call " vascular bonds, " or even through mere intercellular or segmenta- 

 tion cavities. 



Let us look and see if we have any evidence for this theory of growth 

 in the later phenomena of maturation of the fish ovum. We know that 

 in the later stages of egg-development the protoplasm is to a great 

 extent superficial or peripheral in position. In the process of germ- 

 formation at one pole we in reality behold neither more nor less than 

 a growth, ui)building, or aggregation of the protoplasm from the sur- 

 face of the vitellus, or even from its interior, to form a germ-cell which 

 will be segmented oil" at the time of the first true cleavage. That cleav- 

 age will, however, leave behind a modicum of germinal protoplasm sur- 

 rounding or even insinuated into the yelk in strands, which will be the 

 efficient cause of the transformation of what remains of the deutoplasm 

 into a lorm fitted to enter into the further development of the embryo 

 as a plasmine-like substance or as veritable nucleated protroplasmic 

 bodies. These later phenomena, after development has progressed to a 

 definite stage, are everj^ whit as much to be considered phenomena of 

 growth as if the embryo were already feeding. The incorporation of a 

 food yelk seems to a certain extent almost like the process of digestion 

 in an amoeba, with oidy this difi'erence, that the deutoplasm cannot pro])- 

 erl}' be called dead matter, like the food of the amoeba, but protoi)lasm 

 in an inert or quiescent state. The i)arallel does not stop here, however, 

 for \\ e actually have fiuid spaces formed in many embryos around or 

 partially around the yelk, or in closed cavities surrounded by cells, just 



