224 Dr. G. C. Wallicli on the Extent and Causes of 



test is so far asymmetrical (and in this respect differs in cha- 

 racter from all those previously described, with the exception of 

 D. spiralis), that a section passing transversely through the 

 apertural plane viould form tvv^o very unequal portions. Never- 

 theless it can be shown that even here there is no valid ground 

 for assuming specific distinctness, or that the figure of the test 

 is not determined by extrinsic conditions. 



I have already explained how readily the slightest inequality in 

 the distribution of the mineral particles in the earliest state of the 

 test may cause it to assume an oblique figure, and that such figure 

 may, without any improbability, be also imparted to it, in the 

 young state and (as will hereafter be seen) at subsequent periods of 

 its history, by the action of running water, — D. spiralis affording 

 a marked example, in which the latter agency would seem to pro- 

 duce the effect in the most signal degree*. Now, although I 

 have heretofore been unable to satisfy myself that in habitats 

 in which there is a current, and in those in which there is none, 

 the asymmetrically and symmetrically shaped tests respectively 

 predominate to such an extent as to leave no room for doubt as 

 to the efficacy of the second cause I have suggested, from such 

 observations as I have made on the character of the spots from 

 which I have obtained Difflugice, I cannot help believing that 

 this is the case. And whilst I confine myself, at present, to 

 stating the matter suggestively, I may mention that in three 

 localities very widely removed from each other, and in which 

 the nature of the land is itself evidence of the liability or other- 

 wise to a constantly running state of the water, this preponderance 

 seems at all events undeniable. Thus on the borders of the 

 Sunderbunds, where the whole country is a vast swamp, there 

 are pools perpetually fed with fresh supplies of water, but only 

 subject to currents during the inundations occurring during 

 the rainy season. From such pools I obtained the most highly 

 developed varieties of the symmetrical Difflugian tests I have ever 

 seen, namely, D. corona and D. lageniformis (figs. 2 c and 4 c), — 

 the oblique series and even D, spiralis being, however, moderately 

 represented. At Goodhaab again, in West Greenland, I obtained 

 material from pools occurring along the course of somewhat 

 precipitous valleys, and accordingly under the constant action 

 of the mountain streams by which these waters were supplied. 



* Fig. 3 V gives the outline of a normal test of the pyriform variety 

 of Diffiugia proteiformis, whilst the dotted outline represents an ideal view 

 of the same test made to assume a re^orf-shape by a force acting upon 

 it laterally and in one direction. It is an instructive fact that the semicir- 

 cular fold observable in the neck of this test and the neck of an ordinary 

 retort are precisely similar ; indeed the derivation of the word " retort " 

 explains the manner by which the curvature is effected in both cases. 



