4 BULLETIN 104, UNITED STATES NATIONAL MUSEUM. 



constituents are inconspicuous as in Ammodiscus. This cement 

 gives the characteristic color to many of the species of the family. 



Siliceous. — In a few species there seems to be a siliceous cement, as 

 it is unaffected by acids. Many species either secrete or collect 

 fine amorphous siliceous material which is used in the building of 

 the test wall. 



Selective Power in the I'Viumatiox of thic Test. 



No ap'parent selection. — A number of species, including those of 

 the genus AstrorMsa, simply consolidate more or less firmly the 

 material of the ocean bottom, mud, sand grains, other foraminifera, 

 sponge spicules, etc., indiscriminately into more or less regular tests, 

 the outside usually friable, the inner portion commonly firmer. In 

 such tests as these there seems to be no attempt at any selection, the 

 purpose seeming to be to form a somewhat hard protection to the 

 protoplasmic body. 



General selection. — Various groups of the arenaceous foraminifera 

 have some power of selection in that they take some general consti- 

 tuent of the bottom. For instance, RJiahdammina usually in its 

 various species uses sand grains or occasionally spicules. This seems 

 to be mainly a case of leaving out one element at the expense of 

 another. Fragments of the harder materials are taken instead of 

 the softer mud or, as in ths case of Cntliionina, taking the finer material 

 and discarding the coarser. As there is no particular power shown in 

 the fitting of these particular groups of material in any definite way 

 except in the matter of the smoothness of finisli of the exterior or 

 interior surfaces, the selection can not compare with that which is 

 found in the next group. 



Specific selection. — In a few cases the various species seem to have 

 a gi-eat power of selection of the material of the test and in the arrange- 

 ment of tlie particles which have been selected. The genus Psam- 

 mosphaera, building a generally rounded or irregular test with a 

 single cavity and no definite aperture, has in the various North 

 Atlantic species a great power of specific selection and arrangement. 



The common P. fusca uses only sand grains, cementing them 

 firmly together, often with a lighter colored cement. Off the coast 

 of the Carolinas specimens are abundant w^hich have taken only 

 black grains, although other colored ones are present as well in the 

 bottom material. The size is not definite and often in sm^aller 

 specimens the whole of one side will be formed by a single large grain. 



P. parva has a habit of building a test of sand grains of much 

 more even size and usually adds to the test a single large acerose 

 sponge spicule wliich is built into the wall and projects on either side 

 often to a distance as great as the diameter of the test itself. That 

 this is entirely accidental can not be held, for the specimens without 



