NATURAL SCIENCES OF PHILADELPHIA. 11 



been accustomed to regard as that species, clearly runs into the typical iV. 

 rhomboides, of which it is most probably a colletonemoid (sporangial) offset. 



Two of the numerous varieties of this species, seem to be permanently 

 distinct more so, in fact, than in most permanent varieties although con- 

 nected by intermediate forms with the ordinary type. 



iV. rhomboides, (Ehr.) V. nearly quadrangular, striae, faint, parallel 85 in. 

 001". Length -0022", to 0037". (Smith). 



Sporangial Varieties, (1). (a). " Frustule large, V. rhomboid to lanceolate 

 median line double, presenting at the terminal nodules a peculiar arrangement, 

 somewhat similar to the " porte crayon" of Dr. Greville, as it exists in If. 

 Lewisiana only much less obvious." (pi. ii., fig. 11). 



The transverse striae in this form, are about, 60 in -001", "the longitudinal, 

 about, 45 in -001"." Length variable. 



Hab. In nearly all fossil and recent northern deposits. 



(2). (b). N. diaphana? (Ehr.) V. lanceolate extremities slightly pro- 

 duced obtuse, median line thicker than in var. a, terminating in obtuse, 

 rounded, nodular expansions rarely attenuated laterally striae transverse , 

 about 50 a 55 in -00 L". Longitudinal, coarser and more wavy than in var. a, 

 more or less indistinct for some distance round the central nodule. 



Hab. Bemis Lake, abundant, Saco pond, Wolfboro, and other New Eng- 

 land deposits. It is not so common as var. a. (pi. ii., fig, 10). 



The former of these, is probably the ordinary sporangial variety of iV. 

 rhomboides ; the latter, I have found more rarely in gatherings along the Saco 

 Valley, and I think, elsewhere in New England I have been struck with 

 the rarity of both these varieties, particularly, of var. b. on foreign slides.* 



Var. a, sometimes approaches var. b in outline ; but I have never seen the 

 latter of a quadrangular or rhomboid shape. Occasionally the terminal 

 nodule undergoes a trifling modification, becoming slightly indented at the 

 rounded corners. This may, perhaps, be a compound variety. 



Figured at pi. ii., fig. 21, is what I believe to be N. Carassius, (Ehr)., (iV. 

 cocconeiformis, (Smith). (Gregory's new British sp. Mic. Jour. vol. iv. pi. L, 

 fig. 22). This is a rare American species, usually fossil. The strias are very 

 hard to resolve. Some of the varieties of this diatom, suggest an affinity to 

 2V. rhomboides ; but not sufficiently marked to warrant notice. 



" Navicula Sillimanorum," (Ehr.) An exaggerated variety of this species, I 

 have figured, (pi. ii., fig 8). I nave found it only in the Wolfboro mud. The 

 typical form of Ehrenberg, occurs in some of the N. Hampshire deposits. It 

 is a singularly compact and beautiful diatom more nearly allied to N. 

 nobilis, than to N. tabellaria as is shown by the relations of the striae to the 

 marginal line, "which according to the rule before mentioned, indicate the in- 

 tensification of a, primary undulate outline. 



Navicula gastrum, (Ehr.)? fig. 17, pi. ii). Rare in the Wolfboro mud. I 

 have not come across it elsewhere. The striation is more or less irregularly 

 punctate around the median nodule, and along the median line, as in N. 

 granulata. (Bailey). The striae are radiant otherwise, it might be set down 

 as a degenerate variety of N. firma. 



11 Naviccla placenta," Ehr. fig. 4, pi. ii. N. apiculata, Greg. (Mic. Jour., vol. 

 iv., pi. L, fig. 13) N. rostellum. (Smith). This little species is ventricose or 

 elliptic, with a nipple-like projection, at each apex. It corresponds per- 

 fectly in outline, with Dr. Gregory's figure; but the striation is so peculiar in 

 its arrangement, that I have thought it best to figure the valve. The stria; 

 are of two kinds. (1). transverse, sharp, radiant, close. (2). obliquely curved 



(*Tlie slides coming from abroad, labelled "Amician test," are, I believe, prepared from 

 material obtained from this country.) 



1865.] 



