122 MISS M. E. J. CHANDLEB ON THE [yol. lxxix 7 



figs. 9-11). A slight modification of it is seen in Stratiotes 

 tuberculatus, where the raphe enters the seed-cavity slightly 

 below the apex, in such wise that the diagonal is a little shorter 

 and more abrupt than in S. kaltennordheimensis and S. aloides 

 (PI. VI, figs. 12-14, 16). 



The cells of the interior of the keel. — In Headon and 

 Bembridge species these are very complex; interlocking digitations 

 are developed across the keel from seed-cavity to external edge. 

 The digitate cells are convolute, contorted or puckered, so that 

 complicated patterns are produced (PI. VI, figs. 23 & 24). In 

 Oligocene and Lower Miocene seeds, the digitate cells are elongate- 

 parallel to the long axis of the seed, and may still be traced 

 across the width of the keel (PL VI, figs. 25-27). 8. tuberculatus 

 (Upper Miocene, Lower Pliocene) is interesting, for, while it retains 

 the sturdiness and the rugose ornamented testa of the older species, 

 in the cell- structure of its keel it forms an important link between 

 these older species (which some would still place in the genus 

 Folliculites') and the Pleistocene and recent species. Thus the 

 digitate cells are found at the inner edge of the keel ; but they 

 give place to non-digitate, elongate, parallel- sided cells on the 

 keel itself, a character which is also seen in S. intermedins and 

 S. aloides (PL VI, figs. 28-31). In keel and micropyle S. tuber- 

 culatus also approaches these more recent forms (p. 121), yet 

 the fact of its belonging to the same genus as S. Jcaltennord- 

 lieimensis cannot reasonably be questioned (see PL V, figs. 15-17 

 and PL VI, figs. 9-14). 



The Pleistocene and Preglacial S. aloides furnishes a complete 

 evolutionary series, in which gradual changes can be traced from 

 faintly tuberculate individuals that recall S. intermedia to per- 

 fectly smooth examples indistinguishable from the type (recent) 

 S. aloides. These tuberculate and smooth forms are found side 

 by side in the same bed, which may furnish a graded series of 

 specimens illustrative of the passage from one to the other. The 

 evolutionary series is so completely set forth that it is impossible 

 to draw a dividing-line between those provided with tubercles and 

 those (like the recent S. aloides) lacking tubercles. An earlier 

 stage is probably represented by the closely-allied S. intermedins, 

 in which the rugosity of the testa is a constant character, and 

 it is the occurrence on the one hand of this rugose Preglacial fossil,, 

 and on the other of a recent smooth seed, which brings out the 

 significance of the association of smoother and rougher individuals 

 in Pleistocene deposits (see PL V, figs. 18-23). 



Thus the Tertiary and Quaternary Eras furnish a connected 

 series, the members of which constitute links in an evolutionary 

 chain, while sufficient evidence remains to indicate something of 

 the general lines along which changes have taken place. 



