NORTH AMERICAN GRAPTOLITES SI 
The periderm consists of three, rarely only two,’ longitudinal series of 
meshes of a subrhomboidal shape which alternate in adjacent rows, and give 
off from the middle points of the meshes of the outer rows (the rows along 
the ventral margin) short, stout spines which are the mouth ledges crushed 
V-shape. The relation of the three rows of peridermal meshes to the 
skeleton is not known. ‘Uhe parietal ledges form the upper and lower bor- 
ders of the meshes, and are deflected inwards (7. ¢., into the intra-polyparial 
space) to their virgular connection at the inner borders of the outer rows of 
meshes (?), The meshes are covered by a membrane which is markedly 
thinner in the center of the mesh. 
The structure is therefore in substantial agreement with that observed by 
Holm in Retiolites and Stomatograptus, the latter of which, Reteograptus,, 
seems particularly to resemble. 
DICTYONEMA HALL, 1851. 
Am. Jour. Sci., XI., p. 4o1. Type D. retzforme (Hall). 
Like Mr. Holm? I think the taxonomic condition in this genus very unsatis- 
factory. While there seems no possibility of denying to D. flabelliforme 
the possession of a true sicula, certain other species are certainly non-sicu- 
late. The extraordinary vertical range is also, as Mr. Holm remarks, good 
reason for suspicion of the generic references. When I first studied Des- 
mograptus macrodictyon, | thought Mr. Hopkinson’s genus was a first step in. 
the establishment of a natural series of cleavage planes in the genus, espe- 
cially as both species occur at equivalent horizons. But on subsequently 
studying D. devonicus | found that on no characters now predicated of Des-. 
mograptus could this species be denied admission to it. So that we only have 
two wide-ranging genera instead of one such genus. Another explanation. 
(one which has been suggested before, and one which, though I at first could 
not favor, I incline now to think not impossible) is that the characters (form,,. 
dimensions of mesh, thickness, etc. of branches) on which (being dependent 
on conditions of fossilization) we have to rely are really of very subordi- 
nate biologic value. And as Mr. Holm says there is no chance of a rational. 
subdivision until we know more of the basal end. 
Dictyonema, cf. neenah Hall. 
Rep. Progr. Supt. Geol. Surv. Wisc., 1861, p. 7. 
A single specimen, perhaps, referable to this species is found in the col- 
lection. It shows a considerable portion of the network, but not the proximal 
*Several specimens show two series*below and three above, the interpolated 
middle series being wedge-shaped (7. e., narrower and less perfectly developed) below. 
The specimens showing the periderm are mzch larger than those showing the skele- 
ton, and though plainly congeneric, may not be cospecific. 
? Bih. t. k. Svensk. Akad. Handl., 1890, XVI., Afd. IV., No. 7. 
