Thurammina papillata Brady : a Study in Variation. 541 



bodies found in other shells, and we have no doubt that they 

 represent the initial stages of reproduction. In other specimens it 

 will be observed that a limited number of the gemmules have 

 increased in size at the expense of the rest, and have developed into 

 small spherules of chitin, more or less filling the cavity of the 

 specimen, according to their number, which varies in observed 

 specimens from a single chitinous spherule to a crowded mass 

 filling the central cavity (PL XXIX, fig. 11). Sometimes a single 

 large chitinous sphere will be found occupying the greater portion 

 of the central cavity. At this stage personal observation necessarily 

 •ceases, and we must assume that the chitinous spheres are set free 

 by the resorption of the parent test and commence an independent 

 ■existence. Wliile in the test the spherules are always chitinous, 

 and present no aperture or sign of papillar formation, nor, as a 

 rule, are any protoplasmic contents visible through the chitin, which 

 varies in colour from almost glassy transparency to a deep violet 

 or purple colour. The colour may be partly due to the thickness 

 of the chitinous wall, which is very variable, ranging from a mere 

 film, so thin as to be collapsible, and to give diffraction colour 

 effects, to a wall having an appreciable thickness, as seen in optical 

 section. 



The life-history of the spherules after their liberation is at 

 present obscure. As already stated, we think it not improbable 

 that such spherules will be discovered in pelagic gatherings, and 

 so furnish evidence that Thuraminina, like its isomorph Orhulina, 

 passes the initial thin-walled stages of its career at, or near, the 

 surface, before sinking to the bottom to complete its life-history 

 with an acwlutinated shell-wall. The formation of gemm.ules and 

 spherules has been observed by us in T. papillata, T. albicans, 

 and T. cariosa, and there can be little doubt that it extends to 

 all the other varieties. 



The spherules range in size between 0" 01 mm., the smallest 

 measured, and • 2 the largest. The smallest free living chitinous 

 spheres measured were 0-2 mm., but smaller collapsed specimens 

 have been observed. The smallest arenaceous sphere measured was 

 only 0*12 mm. 



But although our pelagic theory must await confirmation, the 

 fact remains that such chitinous spherules (PI. XXVI, fig. 1) are 

 abundant in the fine floated material from the " Goldseeker " 

 stations where Thuraimnina occurs in any numbers, and their 

 subsequent life-history can be followed with some certainty. 

 Continuing to increase in size, they gradually acquire oral papillte, 

 often marked at the outset by surface perforations only ; they 

 secrete investing walls of varying characters, although this 

 arenaceous investment is often postponed until the organism has 

 acquired a comparatively mature condition and assumed those 

 protean shapes which baffle taxonomy. This early and continued. 



