Ski-tkmber 11, 1903.] 



SCIENCE. 



339 



respects, C'romeria appears to agree with 

 Kneria. There are, however, two notable 

 discrepancies. 



Kneria has ' the margin of the upper jaw 

 formed by the iutermaxillaries,' according to 

 Dr. Giinthcr, while in Cromeria Dr. Swinner- 

 ton found that ' both premaxilla and maxilla 

 are small and edentulous,' and that "the latter 

 overlaps the former dorsally and enters largely 

 into the formation of the gap.' In view of 

 the very small size of the fishes and the 

 ambig\ious character of the mouth parts, the 

 apparent difference may be rather nominal 

 than real. 



Kneria has its body ' covered with very 

 small cycloid scales,' while Cromeria lias the 

 body ' naked.' Further, Kneria has a normal 

 tail, while Cromeria has a membranous ex- 

 tension from the caudal above and below. It 

 is possible that both of these characters may 

 be indicative of immaturity (as analogous 

 ones are in some other fishes) but it may be 

 better for the present to assume that the two 

 genera Kneria and Cromeria are distinct; that 

 they are related there is little doubt. 



The family, as represented by Cromeria, is 

 so remarkably distinguished by osteological 

 characters, especially the attachment of ' the 

 greatly elongated arm of a bifurcated post- 

 temporal to the supra-occipital bone,' that 

 it should be isolated as the representative of 

 a peculiar superfamily — Ivnerioidea. As 

 Boulenger and Swinnerton have indicated, the 

 scapular arch being destitute of a meso- 

 coracoid, the group may provisionally be asso- 

 ciated in the same great group as the pikes 

 and killiefishes — Haplomi- — or, perhaps better, 

 in the group Iniomi, inasmuch as the family 

 agrees with those fishes in their technical char- 

 acters. Whether such an association would be 

 natural will be for the future to determine. 

 Theo. Gill. 



the flora of the serpentine barrens of 

 southeast pennsylvania. 



Parts of Montgomery, Delaware, Chester 

 and Lancaster Counties, Pennsylvania, are 

 noted from a geologic standpoint for the 

 presence of outcrops of serpentine rock. This 

 rock formation is confined to the district 



southwest of the Schuylkill River, extending 

 in a somewhat southwestvvard direction into 

 ilaryland, near the lower Susquehanna Kiver. 

 The largest outcrops near Philadelphia occur 

 iu the neighborhood of Lima, Delaware 

 County, at Xewtown Square, at places north 

 and southwest of West Chester, while isolated 

 patches exist south of Bryu Mawr and north- 

 west of Media. There seems no doubt but 

 that all the serpentines in southeast Pennsyl- 

 vania are altered igneous rocks, either pyr- 

 oxeuites or peridotites.* 



The flora of the serpentine exposures, which 

 are always more or less barren in appearance, 

 is peculiar. The eye of the botanist, or of 

 the observant layman, is at once arrested by 

 the association of the characteristic species 

 which make up the serpentine flora, because 

 it is sharply demarcated from the flora of the 

 surrounding country. The botanist can iden- 

 tify the serpentine areas, where the rock is 

 covered by a shallow soil, by the vegetation 

 alone, for the species which are character 

 plants, although occurring elsewhere in the 

 region, are here grouped together in such a 

 manner and in such number, as to delimit 

 sharply these areas from the surrounding 

 country. The serpentine plants taken to- 

 gether, therefore, form islands set down in a 

 sea of other vegetation with a boundary as 

 well characterized as the shore of aYi oceanic 

 island, and with tension lines induced by the 

 struggle for existence as sharply drawn as the 

 shore line against which the storm waves beat. 

 This sharp delimitation of the boundaries of 

 the serpentine areas is emphasized by the fact 

 that these areas are rarely cultivated, but are 

 surrounded by rich cultivable land from which 

 the original vegetation has been removed by 

 man. Many of the plants found on the ser- 

 pentines have survived, therefore, such vicissi- 

 tudes and have persisted on the barrens, while 

 the same species have been exterminated in 

 the cleared land. This fact, however, does 

 not militate against the unique character of 

 the serpentine flora, because the forest, which 

 exists on soils other than the serpentine, is 



•Rand, Theodore D., 'Notes on the Geology of 

 Southeastern Pennsylvania,' Proc. Acad. ATaf. 

 Set. Phila., 1900, p. 305. 



