630 INVERTEBRATA CHAP. 



EXPERIMENTAL EMBRYOLOGY OF SIMPLE ASCIDIANS 



We must now describe some illuminating experiments which 

 Conklin (1905, 1906) performed on the egg of Cynthia. The course 

 of the ordinary development of this form strongly suggests the view 

 that the different coloured cytoplasms, found in the unsegmented 

 egg, are definite organ -forming substances ; hut, of course, this 

 is not necessarily so, and attempts to correlate coloured substances 

 in the eggs of the sea-urchin, Stronaylocentrotus, with organ-forming 

 substances have resulted in failure. 



Conklin found that the ordinary means of isolating the blastomeres 

 of segmenting eggs were not applicable to Cynthia. If eggs were 

 deprived of their membranes by shaking they perished ; and the 

 same result followed if an attempt were made to cut them into two 

 1 >y means of a sharp scalpel. If, however, the segmenting eggs were 

 "spurted," i.e. were sucked violently up in a pipette and blown out with 

 equal violence a considerable number of times, then, frequently, part 

 of the egg, i.e. one or more blastomeres were killed, but the survivors 

 continued their development and in some cases larvae emerged. 



If one of the first two blastomeres be killed in this way the 

 survivor segments exactly as if its dead sister were still present. In 

 this way half-gastrulae are formed, which are like normal gastrulae 

 laid open lengthwise ; the edge of the cup is bordered by a quarter 

 circle of notochordal cells in front, and by a quarter circle 

 of mesenchyme and muscle cells behind. When these gastrulae 

 become larvae the missing parts are not regenerated. There are 

 muscle cells and mesenchyme cells only on one side of the body, and 

 though the ectoderm cells extend round so as to cover in what would 

 lie the naked side of the notochord, yet the notochord itself is only 

 formed from one row of cells, not two as in the normal larva (Fig. 

 458, B). At the posterior extremity of the tail the muscles of the 

 injured side do, to a limited extent, grow on to the uninjured side. 



If the egg is " spurted " in the 4-cell stage, any one of the four 

 blastomeres may be killed, or the two anterior or the two posterior ; 

 or only one blastomere may remain alive, and this latter may be any 

 of the four. In what we may call anterior half-larvae, neural plate 

 and notochordal cells are formed, but there is no trace of muscle cells 

 or of mesenchyme; the neural plate is never enrolled so as to form a 

 nerve tube, the notochordal cells form only a plate, and invagination 

 does not take place. Jn posterior half- larvae, muscle and mesenchyme 

 cells are well formed, but there is no trace of neural plate or noto- 

 chord, and the endoderm forms a solid mass ; and, since there is no 

 notochord, the muscle cells meet in the mid-dorsal line. The mesen- 

 chyme cells form two masses beneath the muscle cells, and are 

 separated by two rows of endoderm cells. The embryo remains 

 rounded and shows no trace of the tadpole form. Quarter embryos 

 are still more imperfect, but what develops out of them corresponds 



