436 Mr. W. G. Tliaewood oyi 



been received from Prof. W. C Sl'Intosh in a collection of 

 specimens taken over from bim by the Museum. In the 

 letter that accompanied the collection Prof. M'Intosli 

 •writes : — " There are also sorac annelids, a few o£ which 

 require working up, a CephaJodiscus? from Kerguelen, and 

 sundry other things." Tiie original 'Challenger' label on 

 the bottle bears the words " Kerguelen, 20-60 fms." in ink, 

 and in pencil, in Prof. M'Intosh's handwriting, " Cephalo- 

 disciis? and a curious Polyzoan." The Polyzoan, which is 

 attached to the coenoecial tubes of the Cephalodiscus, is, I am 

 informed by Sir Sidney Hanner, probably Bcania ma- 

 geRanica. 



This second buttle has a capacity of 70 c.c. oidy, and the 

 eight fiagments that it contains are all small. Judging from 

 tlie difference in the records of the depth — 20-()0 fath. on 

 tliis bottle, and 25 f.ith. on the larger bottle — it would seem 

 that tiie two lots of material did not come up in the same 

 dredging; indeed, it is possible that they were not obtained 

 on the same day, for the ' Ciuillenger ' remained off Kerguelen 

 Island for three weeks. 



Of the two best pieces in the larger bottle, shown on 

 PI. XII., the larger (B) measures about 57x44x30 mm. 

 The coenoecial tubes vary from 20 to 45 mm. in length, and 

 have a uniform internal diameter of 1*0 mm. The external 

 diameter of the upper parts of the tubes that stand out 

 freely, and are not connected by common coenoicial substance, 

 is 1'6 or 1*7 mm. Some of the tubes are bulbous at tiieir 

 lower, blind ends, the greatest diameter observed in a bulb 

 being 1*4 mm. The long tubes show a few concavo-convex 

 septa, irregularly disposed, but confined mainly to the lower 

 ends. The extent to which the free part of a tube stands out 

 from the common coenoecial substance varies considerably, 

 mostly within the limits of 10 and 30 mm. Sand-grains 

 occur embedded in the walls of the tubes and in the common 

 coenoecial substance. 



Tiie upper ends of the tubes differ from those of Cephalo- 

 discus densus dredged by the ' Terra Nova ' in occasionally 

 showing a lateral lip. The majority of the tubes resemble 

 those represented in the accompanying text-figure, a and b, 

 and have the terminal ostium transverse or oblique, without 

 any marked lateral extension ; in this respect tliC}' resemble 

 the tubes of the 'Terra Nova' material (i6, p. 42, text-fig. 4), 

 although there is a larger proportion of strongly oblique ostia 

 than in the latter. But some of the tubes have a laterally 

 extended ostium (text-fig., c), or a tongue-shaped lateral 

 ]»rocess (g), or even a fun"nel-shai)ed ostium (d). A peculiar 

 feature, re[)resented in the text-figure, e, f, g, sugg(!sts that in 



