ME. W. B. S. BENHAM ON THE TESTIS OF LIMULUS. 365 



the pericardium ; the ducts theu become excessively fine, and the network, at any rate 

 the sides of it, ends at about the xi. appendage. The network was here incomplete, as I 

 had removed the portion A previously to noticing these fine posterior ducts. 



The lateral network of the right side is doubtless similar to that on the left ; but I did 

 not follow it throughout its extent. 



Thus there is a continuous network of ducts more or less all over the body, lying, in 

 the median line, below the pericardium and above the intestine; this median network 

 runs throughout the abdomen and thorax, though the ducts are larger more anteriorly. 

 This corresponds to Prof. Owen's "anterior and posterior median lohe" of the ovary; 

 the lateral lobes or branches of this are represented by the lateral networks outside the 

 coxo-tergal muscles of the thorax. 



That the testicular network does thus extend everywhere is shown by the fact that, 

 when the carapace is cut off, a white viscous fluid flows out, which consists of blood mixed 

 with spermatozoa ; this soon coagulates. 



The spermatozoa consist of a small rounded head about two iuch in diameter, with a 

 flattened wide " neck " and a long mobile tail (fig. 1). The mobility of the spermatozoa 

 of Limulus was discovered and recorded by Prof. Lankester in 1881 (Quart. Journ. Micr. 

 Science, vol. xxi.). 



The sperm-sacs are seated in groups upon the sides of the ducts ; two or three, some- 

 times more, are in connexion with one another, one of which opens into the duct (figs. 3 

 & 5). These sacs are subglobular in shape, and are surrounded by a delicate membrane. 

 These are filled with groups of spermatozoa without tails, their further development 

 apparently taking place in the ducts as they approach the aperture. Sometimes sperm- 

 sacs are seen at some distance from a duct, and I could, in many cases, trace no ductule 

 from them (as at B, fig. 5). Perhaps it is here that the sperm-cells are produced, and 

 by continual enlargement of the sperm-sacs, by subdivision of the sperm-cells, and 

 stretching of membrane round them they reach the ducts. 



All these structures, as well as the caeca of the great gastric gland, are packed by a 

 fibrous connective tissue, which is very loose (fig. 5, C). 



The wall of the duct (fig. 6 ) is made up of interlaced tendinous-looking fibres ; by 

 tendinous I mean that they are more or less highly refractive and parallel to one another. 

 These are arranged longitudinally, circularly, and obliquely, while the inner surface is 

 lined by columnar cells. The duct is often filled with the unripe spermatozoa, which, 

 here and there, have aggregated into rounded masses near the walls (fig. 5, S). 



Significance of the above observations. — The importance of the observations thus 

 recorded depends upon the fact that in no crustacean do the ducts of the generative 

 glands form a network, whereas in the Scorpion (and other Arachnids), as in Limulus, 

 they do. Hitherto the fact that the ovarian ducts of Limulus form a reticulum was 

 the only information which we possessed as to the form of the generative organs in 

 Limulus. The observations recorded in the present paper show that the duct of the 

 male organ in Limulus also has the form of a reticulum. Thus a new and important 

 confirmation is afforded to the proposition maintained by Prof. Lankester in his essay 



