108 ME. T. SCOTT ON ENTOMOSTEACA 



January 10th (day tow-netting). Lat. 1° 55' 5" N., long. 5° 55' 5" E., 460 fathoms, 

 January 22nd (day tow-netthig). Station 23, 10, 185, 235 fathoms, February 5th, &c. 

 (day tow-nettings, and a surface night tow-netting). 



This interesting species was obtained in 81 tow-nettings, 16 of these being surface and 

 15 under-surface collections. One of the surface and 9 of the under-surface tow- 

 nettings were day collections, while 15 surface and 6 under-surface were night collections, 

 as shown in the formula : — 



r 1 day collection. 

 rl6 surface | 15 uigi^t coUections. 



Tow-nettings 31^ . 9 day ditto. 



(^ lo under-surface < •' 



I 6 night ditto. 



The under-surface tow-nettings ranged from 3 to 460 fathoms. 



Clytemnestra rosfrata was of more or less frequent occurrence in nearly all the tow- 

 nettings in which it w^as observed. Siiecimens carrying ova were not uncommon in some 

 of the collections. 



The form here described, including its real and supposed varieties, has been the 



subject of a good deal of misunderstanding and controversy, due in great part to the 



meagreness and insufficiency of Dana's description and figures. Nevertheless, after 



having carefully dissected and examined a large number of specimens, I have no doubt 



whatever that the species under consideration belongs to Dana's Clytemnestra, and I 



hesitate to ascribe it to his Clytemnestra scutellata only because of the difference in 



number of the joints of the inner branches of the first pair of swimming-feet, which in 



C. scutellata, as described by Dana, are 3-jointcd, but which in the ' Buccaneer' specimens 



are 1-jointed; and also because of the very marked difference in the form of the dorsal 



aspect between the ' Buccaneer ' specimens and Dana's figure of Clytemnestra scutellata. 



Dana's figure represents the thoracic as decidedly broader than the abdominal part of the 



body, whereas the form of the ' Buccaneer ' specimens is elongate-narrow, with the breadth 



gradually diminishing from the head to the last abdominal segment, and in this 



respect they agree with Goniopsyllus rostratus, Brady, and Goniopelte gracilis, Clans. 



Further, the ' Buccaneer' specimens differ from GoniopsyUus rostratus, as, figured and 



described by Prof. Brady in the ' Challenger ' Report, in two important points : first, the 



posterior antennse of the ' Buccaneer ' specimens possess a rudimentary but yet distinct 



secondary branch bearing two plumose hairs, while in Goniopsyllus rostratus the secondary 



branch is wanting, being represented by a single plumose hair attached to the end of 



the basal joint of the primary branch ; second, the inner branches of the first pair of 



swimming-feet in Goniopjsyllus rostratus are described as 3-jointed, whereas in the 



' Buccaneer ' si)ecimens they are only 1-joiuted, and are so in both the male and female. 



On the other hand, the species described and figured by Prof. Claus as Goniopelte gracilis 



agrees, so far as I can make out, in every essential particular with the ' Buccaneer ' 



specimens, so that the ' Buccaneer ' specimens appear to belong to Goniopelte gracilis, 



Claus, rather than to Goniopsyllus rostratus, Brady. 



In an interesting paper by Prof. Claus in the ' Zoologischcr Anzeiger,' No. 378 



