Royal Microscopical Society. 7 



least proves that colour and situation (in so far as their number 

 in a certain length is concerned) is valueless. I had previously- 

 observed in the guano specimens faint traces of surface markings 

 on the valves between the granules; in the recent specimens they 

 are much more apparent, and if a valve is examined by means of 

 oblique light these spaces will be found to be delicately punctate. 

 I have also been able to satisfy myself of the nature of the large 

 granules, by mounting a broken valve on the edge of a strip of 

 thin glass and examining (with an i) the fracture in profile, the 

 hemispherical elevations were very apparent. The elevations are, 

 however, not solid hemispheres, but hollow, with a minute pore 

 opening into the interior of the frustule; the following diagram 

 represents an ideal section of a valve ; Plate XXI. (lower portion), 



In one of the slides sent are three species of Aulacodiscus in 

 which some of the cells are not filled with balsam ; this eftect may 

 be obtained by transferring the valve into a drop of very stiff balsam, 

 but I have no doubt that sometimes the pore is impervious. 



In the Iquique material I observe many fine frustules and 

 valves of what is supposed to be OmphaJojoelta versicolor of Ehren- 

 berg. The form described by Ehrenberg under that name was 

 found in the " Bermuda earth," and which is now known to be 

 identical with the new Nottingham deposit. 



I have examined many hundreds of slides of the latter, but have 

 never found the form now known as 0. versicolor. I have no 

 doubt but that the true form is only the secondary plate of Helio- 

 pelta, and which is of frequent occurrence in this deposit. 



The species found in the " Monteray eartli," " Mexillones 

 guano," and the Iquique gathering, have the alternate elevations 

 and depressions very conspicuous, and on each side of the elevation 

 a smooth space may be seen, the remainder of the valve (the centre 

 excepted) is marked with minute but distinct granules quincun- 

 cially arranged, on the margin of the elevations, three to seven 

 spines may usually be seen. It frequently happens on transferring 

 a frustule to a drop of water the valves separate, and the secondary 

 plate becomes detached ; this being formed within the cingulum is 

 smaller than the valve, the rays are less elevated, and their margins 

 without spines, the smooth spaces are also wanting, and from a 

 peculiar arrangement of the markings the surface of the disk re- 

 sembles a piece of watered silk, or moire antique ; were it not for 

 these structural differences it might naturally be supposed that these 

 plates were simply newly-formed valves like those of Aulacodiscus 

 formosus. 



The genera Omphalopelta and Heliopelta ought to be merged in 

 Actinoptychus. The following are Kiitzing's specific chai'acters of 

 0. versicolor (species Algarum, p, 133). 0. versicolorYAw. Monats- 



