ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 241 



considerable power of control over its own movements, and is by no 

 means so much at the mercy of the winds or waves. If upset in the 

 water it returns to its original position by bringing its tentacles 

 together over the disk, and throwing up the free edge of the mantle 

 in a given direction, then expanding the tentacles of one side far over 

 in the opposite direction beyond the central part of the disk ; thus, it 

 readily changes the centre of gravity and tilts the overturned disk 

 back again. Medusae are to be found at all stages of development. 



Prof. Agassiz suggests that Porpita is allied to the Hydro- 

 corallinfe, and he bases this suggestion on the possession of the 

 so-called white plate, the peculiar structure of which reminds him of 

 the corallum of Sporadopora, Allopora, and Millepora ; there are large 

 pits, and the whole mass is spongy from being riddled with passages 

 and openings; but there are not, of course, the regular horizontal 

 floors which are seen in Millepora. 



The value of the paper is greatly increased by the twelve plates, 

 two of which give coloured full-sized representations of the two 

 species described. 



Porifera. 



Physiology of Gemmules of Spongillidse* — To this subject, 

 which is now exciting considerable interest. Dr. W. Marshall contri- 

 butes some arguments and observations which should be compared with 

 those given by Dr. Vejdovsky (see below). The wall of the gemmules 

 of Spongilla nitens (as of S. carteri) consists of a system of closed spaces 

 or cells. In S. nitens they form six-sided columns with their long 

 axes tangential to the central mass ; they diminish in size towards the 

 interior of the gemmule ; the outer cells are hollow and in the dry 

 state filled with air, the innermost are solid. These cells are not 

 histological cells, but of cuticular character ; their walls are strongly 

 refractive, and resist combustion stubbornly ; fluoric acid destroys 

 their refractive power and brittleness, so that it appears not impro- 

 bable that they contain a large proportion of silica ; the inner layer 

 of spined spicules is attached to this cellular layer with more firmness 

 than to the subjacent horny layer. ;S^. nitens has an air-space, formed 

 by the chitinous layer, as in S. carteri, which enables the dry gem- 

 mules to float for from 8 to 10 days in water. The elaborate enve- 

 lopes which cover the abundant starch which accompanies the germinal 

 matter provide in the most satisfactory manner for the protection and 

 welfare of this material. Various arguments are advanced in favour 

 of the aerostatic character of the cellular coat of the gemmule, viz. 

 the smalluess and, owing to the great relative development of this 

 layer, the lightness of this body in S. nitens. In the districts where 

 the sponge occurs it must often be left dry by evaporation and the 

 gemmules subsequently set free may be carried long distances by the 

 wind, and eventually germinate if they meet with fresh water again. 

 Thus, of Ehrenberg's figures of organisms found in trade-wind dust, 

 about 24 per cent, refer to sponges, and of these fragments about 16 



* Zool. Auzeig., vi. (1883) pp. 630-1, 618-52. 



