348 H. H. NEWMAN 



is seen in the case of the much-retarded parthenogenetic larvae 

 of Patiria, where blastulae, at first quite devoid of axes of polarity 

 and with no pronounced apical points, undergo gastrulation at 

 points that are relatively basal; for it is at the basal end of normal 

 larvae that invagination of endoderm takes place. In such 

 larvae as that shown in figure 13 the whole periphery is physio- 

 logically basal, and gastrulation occurs anywhere or everywhere, 

 and as many gastrulations occur as there is room for on the 

 surface. 



It is further to be noted that retardation has its effects, whether 

 expressed in twinning or in other ways, only at such stages and 

 in such places when and where some critical change, involving 

 a process of unusual delicacy, is taking place, and where even a 

 slight or temporary retardation would result in a sufficient 

 developmental let-down to bring about physiological isolation 

 of tissues to the right and left of the original apical point. 



The probable causes of twinning in other species 



One of the classic studies of experimental twinning is that of 

 0. Schultze ('94), using the eggs of the frog. He discovered 

 that if he turned the egg upside down when in the two-cell 

 condition there frequently developed twin or double-monster 

 larvae and that the axes of the twin embryos bore no constant 

 relationship to each other. Sometimes the anterior ends pointed 

 in exactly apposite direction; sometimes they were at right 

 angles to each other, and sometimes they bore a mirror-image 

 relation to each other. It is obvious from Schultze's account 

 that inversion of the frog's egg decidedly retards development 

 and necessitates a more or less complete dedifferentiation of 

 the blastomeres, followed by a redifferentiation of a new polarity, 

 involving the formation of a new apical point in each blastomere. 

 If the dedifferentiation were sufficiently thoroughgoing, there 

 resulted complete physiological isolation of the two blastomeres, 

 and separate twins were formed. If dedifferentiation were less 

 complete, double monsters or conjoined twins were produced. 

 The essential fact is not that the eggs were inverted and that 

 specific gravity caused a restratification of formative materials, 



