SUPPLEMENTARY NOTES ON TWINS IN CATTLE. 55 



fusion may take place at a much earlier stage ; and vascular anasto- 

 mosis may date from the time of extra-embryonic circulation. The 

 approximation of the embryos in the common chorion may give 

 some indications of the manner and time of fusion, and the vascu- 

 lar relations will also throw some light on the same questions. 



I have two cases of this kind (Nos. 78 and 97) in which the 

 foetuses are of opposite sex, but in neither case is the modification 

 of the free-martin particularly extreme. They are so alike that 

 the description of one of them (No. 78) will suffice. Fig. 2 is 

 from an oil painting showing the membranes ; the very direct 

 arterial connection between the two umbilical cords will be noted. 

 The two amniotic cavities are so close together that the walls are 

 in contact at one place. The fusion of the two original vesicles 

 was presumably, therefore, side by. side and very early, for all 

 overlapping parts have disappeared without leaving any trace. 



If I were to venture to reconstruct the probable history of this 

 case, I would say that the fusion was probably complete at least 

 by the 10-mm. stage, and vascular anastomosis established at the 

 same time. The reasons for this opinion are the extent to which 

 the membranes of a 10-mm. embryo fill up one horn of the uterus 

 (see Lillie, 1917, p. 389) and the readiness with which fusion of 

 the membranes takes place (specifically in cattle) when they are in 

 contact. If this opinion is correct, the vascular anastomosis in 

 this case dates from long before the period of beginning of sex 

 differentiation. 



The anatomy of the reproductive organs of the free-martin is 

 shown in Figs. 3A and 3B. If these figures be compared with 

 figures of free-martin anatomy given in my 1917 paper and with 

 Figs. 4 and 5, it will be seen that the extent of the modification is 

 less than the average in certain respects. Thus the cornua and 

 corpus uteri are retained in this case, though very rudimentary 

 compared with the normal; in probably the majority of free- 

 martins they are entirely absent (cf. Fig. 5) ; the Wolffian ducts 

 are also more slender than in most free-martins, and the guber- 

 nacula less typically developed (cf. Fig. 4). The gonads are ex- 

 ceptionally minute, which would appear to indicate less than the 

 average stimulation of the male homologues in the original ovary 



