POEAMIXIFEEA OF THE KEEIMBA AECHIPELAGO. 373 



are characteristic of the test (o/Thurammina papillata, to ivhich the organisms were at 

 that time referred hy Earland), are well marked in some specimens, while in others 

 they are entirely absent." 



Comparing this original description with the Kerimba specimens, but slight 

 modifications suggest themselves. The Kerimba tests are quite as variable in their 

 appearance as the British, but at the same time they appear to be divisible into two 

 groups: (i.) a simple test or dome (PL XXXVI. figs. 6 & 8), and (ii.) an irregularly 

 agglomerated mass (figs. 9 & 11). The dome-form is unquestionably the initial stage, 

 and in the smallest specimens it occasionally presents a simple orifice at the top or 

 side of the dome, but it more often has no visible aperture. It was the absence of 

 any aperture in the similar specimens found in the shore-sands of Selsey Bill which 

 led us to refer the organism " with considerable hesitation " to Webbina hemisphcerica. 

 A fresh examination of the Selsey Bill specimens leaves no doubt in our minds that 

 they are identical with the Kerimba individuals, or that the small domed tests represent 

 the initial stage of the large and irregular structure subsequently developed. 



The agglomerate form represents the addition of successive areas to the original 

 "house," the line of separation between these growth-stages being usually marked by a 

 superficial external constriction, which, however, does not correspond with any internal 

 septum. With each addition in size, part of the previously investing wall must be 

 absorbed, as the whole interior of a large agglomei'ate test forms one undivided cavity, 

 although it is often more or less broken up by furcations of the test due to the aberrant 

 growth of the organism. 



The superficial character of the test in the Kerimba specimens is much the same as 

 in the original British examples, but there are some differences in construction, owing 

 to the distinctive character of the material employed, which is shell-sand, coral- 

 fragments, saud-grains, and sponge-spicules. If anything, the fully grown Kerimba 

 specimens are more roughly constructed than the British, and in nearly all cases they 

 are, or have been, attached, free-growing individuals being extremely rare. The 

 Kerimba specimens appear to be much more readily detached from the host than the 

 British, unless this fact is due to their having lived in association with some organism 

 of a perishable nature, such as Zostera, on fragments of which some specimens were 

 found (PL XXXVI. fig. 2). The majority of the detached specimens have not suffered 

 any damage to the delicate and diaphanous chitinous membrane which had been in 

 contact with the host, and from which we have derived its. specific name. 



It seems probable that India diaphana may sometimes exist in a free state on the 

 actual sea-bottom in crannies between large sand-grains, as there are a iew specimens 

 in which both sides of a compressed test exhibit the chitinous film, the adventitious 

 investment of fortuitous material being limited to a circumferential belt (PL XXXVI. 

 fig. 10). Within this chitinous drum (so to speak) the protoplasmic body can often 

 be clearly observed. 



3g2 



