NATURAL SCIENCES OF PHILADELPHIA. 9 



My attention was attracted long since, by the singular likeness, in striation, 

 median nodule, and general character, of Navicula serians to N. follis, (Ehr.), 

 a form not uncommon in northern post glacial deposits. 



Ehrenberg's loose figures of the latter (A r . follis), give so many varieties of 

 outline, and his description is so vague, that in time it seems to have become 

 confounded by the English writers, with Navicula' inflat a, a species having no 

 analogies to it beyond a slight resemblance in outline. 



Indeed it is probable, that Ehrenberg himself, owing to the use of im- 

 perfect microscopes may have been led to associate these two very distinct 

 species, and in distributing specimens may thus have originated the mistake. 



However this may be, it is evident that in describing the form called N. 

 follis, in the Bridgewater peat, (Mass.,) he did mean the one figured as var. b. 

 [cruciform), of N. serians, fig. 5 6. pi. ii.* 



The general resemblance in character, however, would not have sufficed 

 to prove the specific identity of these two forms, viz. N. serians and its var. 

 b. cruciform, had it not been for the discovery of a second variety which 

 bridges over the interspace that figured as N. serians, var. a. fig. 5, a. pi. ii. 

 This variety occurs in two of the White Mountain localities, Bemis Lake, and 

 Echo Lake, along with the typical form and the cruciform, var. b, a strong 

 corroborative evidence beside that afforded by numerous intermediate vari- 

 eties of a common derivation. 



As this is a fact which, if recognized, involves a most important point in 

 its bearing on the laws which determine species, I have drawn with utmost 

 care (fig. 5, a. b. pi. ii.) the valves of what appear to be average specimens 

 of these two varieties. The Bemis Lake' slides of Mr. Samuels, now very 

 difficult to obtain, are very carefully mounted, and afford beautiful illustra- 

 tions of these forms, as well as of Stauroneis Stodderii, n. sp. (Greenleaf,) (fig. 

 6, pi. ii.) The material is quite scarce at the present time, although it could 

 no doubt readily be procured from the original locality. 



(1). "Navicula firma" (Kutz). Large, turgid, oblong, lanceolate, with 

 obtuse, cuneate ends, thick borders, and large median nodule ; striae wanting, 

 or obscure." 



(2). "Varieties." (a), linear oblong, (Navicula iridis.) p. pointed elliptic, 

 (2V. dilatatd). y. cuneate (N. amphigomphus.) 



(3). " Secondary or metamorphic varieties." S. triundulate, (not the same as 

 N. Ilitchcockii). i produced (iV. producta. 2V. affinis). . apiculate. (iV. amphirynchus). 



The specific identity of N. iridis, N. dilatata, N. amphigomphus, and perhaps 

 one or two other reputed species has, I believe, been generally suspected, if 

 not recognized. Certainly, it would be difficult for any one carefully studying 

 the Northern deposits of this country, whether recent or fossil, to resist the 

 conviction, that they really have a common origin. With regard, however, 

 to some of those I have termed secondary or metamorphic, their common deriva- 

 tion from N. firma is not so obvious. I believe them, notwithstanding, to be 

 all varieties of that species, a conclusion I have arrived at, after a careful 

 comparison of gatherings from numerous localities. As, however, a conclusion 

 so formed may be regarded as more a matter of prepossession than admitting of 

 proof, I shall not undertake to illustrate it by elaborate descriptions of these 

 varieties ; but this much may be said, that what has hitherto kept apart many of 

 the best known of them, is simply difference in valvular outline and number of 

 striae. If, hereafter, my view with regard to the unimportance and mutability 

 of these characters be proved correct, the union of some, if not all these 

 reputed species will probably be necessary. 



Taking all the enumerated varieties, there appear to be certain general 



(*I am the more certain of this, from my friend Prof. H. L Smith, of Kenyon College, 0., haTing 

 recently written to me, calling attention to this very fact.) 



1865.] 



