542 Transactions of the Society. 



chitinous condition is, in our opinion, the key to the problem of 

 relationship between the groups, and is responsible for the protean 

 forms which have been figured, especially by Haeusler, under the 

 name " papillata." The flexible chitinous membrane lying in 

 contact with other organisms, and under the influence of many 

 conditions beyond our imagination, responds to the stimuli, 

 collapses and expands according to its environment, and develops 

 abnormal forms of growth. No other explanation will account for 

 the protean forms obtained from the " Goldseeker " material. 



We shall have occasion to refer to the occurrence of associated 

 tests — usually in pairs, but occasionally in groups of three or even 

 four individuals. There is little or no evidence that this pheno- 

 menon is in any way connected with the reproduction of the 

 animal. We agree with Brady that it is merely a case of associa- 

 tion of two or more individuals which have come into contact with 

 each other, and continued their growth side by side as attached 

 specimens. There is no resorption of the walls of the test at the 

 point of contact to provide free transfusion of the protoplasm 

 between the individuals. Neither sections nor balsam-mounted 

 specimens give any evidence of channels of communication between 

 the two tests beyond the normal papillar openings. It is note- 

 worthy that such associated pairs are usually of identical size, 

 though Brady (Ref 12) figures an abnormal association, consisting 

 of one large and two smaller individuals (pi. xxxvi, fig. 15). In 

 the variety T. cariosa it is also to be observed that the associated 

 pairs combine to form a common spongy layer, which gradually 

 fills up the groove separating the two spheres (PL XXTX, fig. 6). 



It is worth recording that, in our experience, the individuals of 

 associated pairs are always identical as regards variety. We 

 have no record of any mixed pairs, although instances occur in 

 which included shells are of a different nature to the outer sphere. 

 These two observations, and the further fact that in our new variety, 

 sordida, the associated individuals of a pair are sometimes bound 

 together by a collajised chitinous membrane, which probably in 

 life contained both individuals as in a common cyst or bag, may, 

 perhaps, be taken as some evidence towards regarding the associa- 

 tion as a sexual process, or even as an alternation of generations. 

 But it would not, in our opinion, be safe to base any deductions 

 upon such isolated records. 



Of quite different significance and much greater variety are the 

 instances of budding. Very few cases of budding have come under 

 our observation, but we figure two instances, in one of which the 

 parent sphere of an irregular haeusleri type is seen in the act of 

 forming several small individuals, which are but loosely attached to 

 the parent (PI. XXIX, fig. 16). In the other a regular castanea 

 sphere has budded a smaller sphere at one pole, and an irregular 

 test at the other (PI. XXIX, fig. 17). 



