NATURAL SCIENCES OF PHILADELPHIA. 387 



midise in abundance. To the former of these alone I propose directing atten- 

 tion, not only on account of the singularity of their departure from established 

 generic types, but also because of certain points of great speculative interest 

 illustrated by their peculiar grouping, and the unique and isolated nature of 

 the deposit in which they occur. 



My friend, Prof. L. W. Bailey, of Frederick, N. B., who is at this time de- 

 voting himself specially to the study of the interesting question of the geo- 

 logical bearing of the Diatomaceae in determining the relative antiquity of 

 fossil beds in the State of Maine and the adjoining country, in a letter to me 

 of some time back, alludes to the fact of the extreme rarity of the genera 

 Synedra and Nitzschia in any of the sub-peat deposits of the northern section 

 of the United States, although, as he adds, both are quite common in surface 

 gatherings. 



His observations I understand to apply to sub-peat (fresh water) deposits 

 alone, and not to the mixed class of alluvial deposits like the Hudson and 

 Delaware river muds, or, in fact, to any foreign post tertiary deposits as those 

 of Mull and Glenshira, which in many respects differ from cotemporaneous 

 beds in this country. 



My own experience agreeing with Prof. Bailey's so far, at least, as regards 

 the rarity of these genera in the sub-peat of this country,* the discovery 

 of the forms contained in the Saco pond was peculiarly interesting to me, as 

 it seemed to foreshadow the abundant introduction of the Synedroid genera, 

 by presenting a series of intermediate types between these and Surirella. 



In advancing my theory of the geological and structural position of these 

 new species, I must premise, however, that it is, in a great measure, founded 

 on a conjectural basis, and from the very nature of the points involved, in- 

 susceptible of positive proof. I shall, therefore, in describing these species, 

 confine myself to a brief notice of the principal facts which appear to me to 

 sustain my views, reserving for a future occasion the consideration of such 

 collateral evidence as I may have derived from other sources, but which the 

 limits of the present paper forbid my here introducing. 



This gathering, as before stated, was derived from the margin of a pond 

 supplied by the springs which feed the river Saco. The position of this pond 

 and its surroundings I am particular in describing, for reasons which will be 

 apparent hereafter. It lies on the eastern side of the Crawford intervale, and 

 is surrounded by other and smaller ponds, which dry up during the summer. 

 It is about 200 feet long, rather less in width and very shallow. At its 

 southern extremity is the outlet of the small streamlet which forms the begin- 

 ing of the Saco river. A few small mountain brooks empty into it laterally, 

 but the main source of its supply appears to be certain springs welling up 

 from underneath the alluvial detritus forming the pond bottom, which over- 

 lies a thin stratum of clay, beneath which is the boulder drift. A sparce 

 growth of grass fringes the northern margin of the pond, and on its southern 

 and western aspect is a boggy patch with a good deal of moss, but this 

 appears to be quite superficial, and at a few inches below the surface is the 

 same alluvial till without much vegetable debris. Similar conditions of 

 soil seem to hold at other points in the valley. It is in this fine and soft mud, 

 at from one to four inches below the surface, beneath and immediately around 

 the waters of the pond, that the silicious remains of the new species are most 

 abundant. 



The gathering is principally remarkable for two points : 

 Firstly. The striking analogy which exists between its species and those of 

 the sub peat deposits of the northern section. 



Secondly. The occurrence of several forms belonging to a known genus 

 Surirella so peculiar and variable in their characters as almost to merit the 



* Synedra ulna, S. radiaris, S. lunaris, S. biceps and two or three speclea of Nitzschia occur 

 sparingly in several sub-peat localities. 



1863.] 



