DOUBLE MONSTROSITY— DEVELOPMENT, ETC. 



together by ventral rather than by lateral union, the formatinn posteriorly of a perfectly single lnnly 

 or tail becomes impossible, since the necessary readjustments of right and left structures in the twin 

 embryos can no longer take place. 



(b) In the second group, the twin centres of embryo-formation are so far apart that there is no 

 compelling influence of the kind described above which would lead to the approximation and union 

 of their growing embryonic axes. Accordingly the twin bodies remain separate, except for the 

 adventitious union supplied by the layers forming the wall of the common yolk-sac. Text-fig. 3 

 illustrates an early example in which the twin embryos are almost at opposite sides of the blastoderm. 

 In the higher animals the corresponding type of doubling may give rise to completely separate 

 unioval or homologous twins. 



As is well known, in the higher animals well-developed unioval twins are always of the same 

 sex, and they tend to show a remarkable degree of resemblance even in minute characteristics. In 

 fishes, however, one of the twins is not infrequently smaller than its neighbour, or otherwise 

 defective, no doubt through injury or want of room for growth. The defect sometimes takes 

 the form of a pure cyclopia. Accordingly we have here a reason 

 for believing that the natural occurrence of cyclopia need not be 

 occasioned by conditions inherent in the fertilised egg, since other- 

 wise both heads ought to be similarly affected. 



There do not seem to be any direct observations enabling one 

 to say what amount of angular distance must separate the twin 

 embryonic rudiments on the egg of the trout in order that the 

 twin bodies may remain apart from one another throughout their 

 whole length. Taking it as a matter of chance at what part of 

 the blastodermic rim each of the twin rudiments appears, one might 

 seek evidence on the question by comparing the relative frequency 

 with which union of the twin bodies and union by the yolk-sac 

 only {anakatadidymus) are found to occur. Windle's figures (272) 

 show two anakatadidymi out of a total of forty-six double forms. 

 My own numbers are seven out of seventy-one. Summation of 

 both gives an average of one in thirteen. Dividing the whole 

 circle of the blastoderm (3G0°) by thirteen we get rather less 

 than 30°. In the trout it would seem accordingly (if one assumes 

 the correctness of the principle on which the calculation is founded) 

 that, in order to remain axially separate, the twin embryonic rudiments must have appeared on 

 the rim of the blastoderm not less than 165°, i.e. 180°— 15", from one another. Probably this 

 estimate of distance is too great. A larger number of instances to average from would likely 

 reduce the figure, but Kopsch's (133-3) experiments seem to show that in any case it is over 90°. 



All double monsters in fishes which are produced in the manner described above will be 

 examples either of anterior duplicity or of union by the yolk-sac only. Further, in all cases the 

 two heads will point in the same direction. Klaussner, however (123, PI. V., Fig. 34), figures an 

 example of union by the yolk-sac only, in which the twin heads are pointing in opposite directions, 

 i.e. the head of the one lies alongside the tail of the other. This seems to be the solitary recorded 

 instance of the kind, and Windle (272), in calling attention to it, notes that it is "difficult to 

 account for by any of the theories now holding the field." Klaussner 's figure is not a perfectly 

 convincing one, but if correct could readily be explained by assuming the presence of two originally 

 separate blastoderms on the parent egg. Such a condition, although it has never been observed 

 in the eggs of fishes, is not unknown in those of other vertebrates. A number of instances are given 

 below. 



The foundation of our knowledge regarding the mode of growth and union of double monster 

 fish embryos was laid by Lereboullet's work (14-3) on the living eggs of the pike (see p. 10). 

 Very early stages (see pp. 9-10) have also been figured for the same species by Bauber (2U0), and for 



Fig. 3.— After Rauber (202 6, Taf. 

 VIII. fig. 14). Egg of trout fourteen 

 days after fertilisation, showing two em- 

 bryonic rudiments, almost, but not quite, 

 opposite to one another on the margin 

 of the blastoderm, a, blastoderm spread- 

 ing over the yolk ; b, yolk uncovered by 

 blastoderm ; c c', the twin embryonic 

 rudiments. 



