THE STRUCTURE AND AFFINITIES OF HERDMANIA CLAVIFORMIS. 251 



opening. Indeed I am unable to make certain of any opening at all in some cases. 

 I have never seen the least suggestion of the passage of an embryo into the atrium, 

 which is always, so far as I have observed, very restricted in size. I am inclined 

 to believe the real method of escape will be found to be by rupture of the uterine 

 and body walls of the zooid. 



The relative size of the gravid uterus and oviduct proper, that is, the duct a 

 short distance in front of the ovary, is shown in Figure 2, o'dt., lit., and in Figure 7, 

 o'dt., and Figure 8, ut. The uterus measures when containing full-grown embryos 

 about .855 millimetre in diameter, while the oviduct proper varies from .095 to .148 

 millimetre. 



That the great distention of the uterine portion is not due to mere passive stretch- 

 ing by the growing embryos is clear from the fact that in reality the wall is consider- 

 ably thicker in the uterus than it is in the empty duct, and further by the fact that 

 the size of the uterus is nearly as great at intervals between two embryos, even where 

 the distance is considerable, as it is where an embryo is lodged. 



At no time do the embryos contract a placental relation with the uterine wall 

 as those of some of the Polyclinidse are reported to do by Salensky ('92). 



The testis consists of a considerable number of lobes each connected with the 

 vas deferens by its own branch (Fig. 2, te.}. A few of the anterior lobes may extend 

 so far forward as to be situated alongside the intestinal loop; but by far the greater 

 number are behind the loop in the postabdomen. The testis may extend back con- 

 siderably farther than the ovary. The usual thing is for the ova and sperm to ripen 

 at different times in the same individual. Only exceptionally have I found a zooid 

 in which both were ripe at the same time. The vas deferens is of necessity very long. 

 Its course is parallel with that of the oviduct, on whose inner side it lies. 



ii. The Mesenchymatous Tissue and the Subepicardiac Space. The entire body 

 mass around the organs of the abdomen and postabdomen is made up of (1) a 

 coarse network of extremely delicate membranes; (2) long brandling nucleated 

 fibres or strands, which might be called connective tissue; and (3) quantities of free 

 cells, varying greatly in form and structure, though probably all modifications of 

 a single kind of cell. 



The membranes are arranged in irregular festoons from the mantle musculature; 

 so that when the spaces are filled with free cells, as is usually the case, they appear 

 on cross-sections of the body disposed in more or less regular columns cut across (Fig. 

 8, cl.}. These columns may often be distinctly seen in the whole animal, and can 

 be traced, also, for long stretches through series of sections. Sometimes they are 

 nearly cylindrical, sometimes distinctly angular, and sometimes much flattened. 



