THE GENUS ASTRORHIZA OF SANDHAL. 223 



seem referable to two fundamental types, which may bej 

 designated respectively the lenticular and the cervicorn. \ In 

 the former the test may be considered as deriving' its shape 

 from a spheroid (fig. 5), which, whilst becoming flattened, 

 sends out irregular digitate prolongations (figs. 6, 8,10, 11, 12), 

 so as very much to resemble the different aspects of an 

 Amoeba, In the latter (figs. 1 and 2) there is nothing that 

 can be called a centrum, but a somewhat flattened stem puts 

 forth irregular branches of the like shape, which often again 

 branch, sometimes widening out into palmate expansions 

 before doing so. Between these there are various inter- 

 mediate gradations, the simpler or more elementary examples 

 of the " cervicorn " type (figs. 4, 7, 13) having an obvious 

 tendency to approach those modifications of the "lenticular" 

 in which the radiations enlarge at the expense of the central 

 disk. When the cavity is laid open, which is best done in 

 the manner recommended in the case of the Lituolce having 

 the like texture (loc. cit.), it is found to be small in pro- 

 portion to the size of the test, the walls of which are extremely 

 thick (figs. 3 and 4). It is in all cases simple and undi- 

 vided, and is occupied by a sarcode-body of a greenish- 

 brown hue. "^^ 



A very remarkable feature in this organism, which in ' 

 other respects might be likened to a Rhabdammitia (loc. cit.) 

 of gigantic size and coarse construction, is the absence of any 

 definite aperture. This I have carefully substantiated by the 

 examination of a considerable number of specimens under 

 various methods of observation; the extremities of the 

 branches or radiations, like every part of the general surface, 

 exhibiting a continuity of the arenaceous layer, save where 

 it has obviously been interrupted by fracture. From the 

 looseness with which the sand-grains are aggregated, how- 

 ever, it seems quite conceivable that the pseudopodial 

 extensions of the sarcode-body may find their way out 

 between them at a7iy points, but more especially at the 

 extremities of the branches or radiations, where the wall is 

 thinner than it is elsewhere. And this deficiency of firmness 

 in the test further suggests that it may undergo enlargement 

 by the extension of its ramifications, the terminal layer of 

 sand-grains being pushed away (so to speak) by the growth , 

 of the sarcode-body. - — / 



Any doubt I might have entertained as to the essential 

 identity of our own types with the Astrorhiza of Sandahl 

 have been removed by the fact that specimens collected by 

 M. Sars, jun., from a depth of 450 fathoms, whilst corre- 

 sponding in every particular with our own gatherings, are 



