418 ^I^i*. II. A. Brtylis 07i Two new 



ihoir membranes ; occasionally, liowever (fig. 1, E.), one or 

 two embryos appear to be already free in utero, and it may 

 be inferred that the species is viviparous. 



There are two large caudal ojand-cells (fig. 1, (7. (?.), 

 situated, one behind the other, just behind the anus, and with 

 long fine ducts passing to tiie adhesive organ at the tip of 

 the tail. 



Among about 100 examples of this species more or less 

 closely examined, not a single male was observed, and I did 

 not notice any individuals whi'h appeared to be herma- 

 phrodite. Hence it seems probable that this species is capable 

 of multiplying parthenogenetically, or at least that the form 

 found in the crab represents a partlienogenetic generation. 

 It is, of course, possible that there is also a free-living bi- 

 sexual generation, 



2. Monhystera carcinicola, sp. n. (Figs. 4-6.) 



Mature ? . Length I'l-l'G mm. (usually 1-1-1-2 mm.). 



a=21-32'7. 



^=5-l-7'2 (usually 6-6-2). 



7=10-5-13-3. 



5 = 8-12-6. 



^ . Length 1-1-1-2 mm. 



a = 24-26-6. 

 y3 = 5-5-6. 

 7=11-5-13. 



Hah. Gill-chambers of Gecarcinus riiricoJa and Gardisoma 

 guanhxnni. 



The first set of measurements given above were taken from 

 ten specimens containing more or less mature female organs, 

 while the second set are based upon the only two specimens 

 which showed distinct hermaphroditism. Of the twelve 

 specimens of this species available, eight came from Gecar- 

 cinus ruricola, and the remaining four from Ourdisoma 

 guanhumi. One hermaphrodite specimen occurred in each 

 host. No important differences were noticed between the 

 two sets, but those from G. ruricola were in somewhat 

 better condition, and for this reason my description will be 

 mainly based upon them. 



This species is, as a rule, distinctly smaller and more 

 slender than M. wUsoni, but at a glance it is not very 

 easy to separate the two. In most respects the anatomy 

 is very similar, but the following are important points of 

 distinction. The head-end (fig. 5) is always slender and 



