· FAUNA OF THE BRAINTREE ARGILLITES.

Wishing to examine typical specimens of Paradoxides Harlani, the writer visited Boston, and through the kindness of Mr. Alpheus Hyatt, curator in charge of the collections of the Boston Society of Natural History, he not only obtained access to the collection from the Braintree argillites, but the loan of such specimens as were wished for study and illustration. Mr. Alexander Agassiz also gave permission to use material in the collections of the Museum of Comparative Zoölogy, and Mr. N. S. Shaler placed his private collection at the writer's disposal. It is owing to these favorable conditions that I am able to present at this time illustrations and descriptions of the fauna of the Braintree argillites.

The first notice of the presence of fossils in the patches of argillite associated with the Quincy granite in the north end of the town of Braintree, Mass., was by the late Mr. William B. Rogers, who called the attention of the members of the Boston Society of Natural History to it when exhibiting specimens of a large trilobite found at Hayward's quarry (Proc. Boston Soc. Nat. Hist., vol. vi, p. 27, 1856). Subsequently Mr. Rogers traced the history of the trilobite described by Mr. Jacob Green as Paradoxides Harlani, in 1834, and showed quite conclusively that it came from Hayward's quarry and was identical with the species found there. Mr. Henry D. Rogers published the best figure of the species yet given, with remarks on its discovery, &c. (Geol. Surv. Penn., vol. ii, p. 816, 1858).

Numerous collectors obtained specimens of Paradoxides Harlani, but it is not until 1861 that we find any notice of other species. Mr. Albert Ordway then states that he had found a fragment of a trilobite similar to that described in the Paradoxides beds of Newfoundland, in association with Paradoxides Bennetti, and which he referred to the genus Ellipsocephalus. He also mentions the discovery of "a distinct fucoidal impression which shows three branches, each about 4 inches long, but not sufficiently well marked to afford any evidence with regard to its nature" (Proc. Boston Soc. Nat. Hist., vol. viii, p. 6, 1862). The small trilobite is probably the same as that subsequently described by Mr. R. P. Whitfield as Arionellus = Agraulos quadrangularis (Bull. Amer. Mus. Nat. Hist., vol. i, p. 147, 1884). Mr. Ordway also published a figure of the head of Paradoxides Harlani, when comparing that species with Paradoxides spinosus, Boeck, which Mr. Barrande considered as identical with P. Harlani (Bull. Geol. Soc. France, vol. xvii, pp. 545-548, 1860). In the year 1863 a restored figure of Paradoxides Harlani, by Mr. F.

B. Meek, was published in Dana's Manual of Geology and repeated in each subsequent edition of the Manual. This very well represents the general characters of the species.

The publication of the second described species by Mr. Whitfield gives a stronger interest to the fauna, which is now increased by the addition of another species of trilobite, Ptychoparia Rogersi, and a species of Pteropod, Hyolithes Shaleri.

In seeking for a fauna in the Cambrian system of North America to compare with that of the Braintree argillites, we are at once directed to the Paradoxides beds of Newfoundland by the almost perfect identity of the leading type of each locality, Paradoxides Harlani and P. Bennetti. I think it has yet to be decided that the two are distinct species. Hyolithes excellens, Billings (Pal. Foss., vol. ii, pt. 1, p. 70, 1874), is very closely related to H. Shaleri, more so than to any other American species, and Agraulos socialis, Billings (loc. cit., p. 71), is of the same type as A. quadrangularis, Whitfield, as shown by figure 1 of plate vii. Ptychoparia Rogersi does not appear to be represented in the Newfoundland Paradoxides beds, unless it be by Ptychoparia (Solenopleura) communis, Billings (loc. cit., p. 72).

Mr. Barrande has shown the strong resemblance between Paradoxides spinosus, of the Bohemian Basin, and P. Harlani; and the Paradoxides beds of Sweden, Bohemia, Wales, Newfoundland, and Braintree have frequently been correlated in a general manner by authors.

Genus HYOLITHES, Eichwald.

HYOLITHES SHALERI, n. sp.

Plate vii, figs. 4 4a-c.

Form an elongate triangular pyramid, slightly arching towards the dorsal side and expanding regularly from the apex towards the aperture. Transverse section midway of the length, semielliptical, with a width twice as great as the height; the lateral angles acute. Ventral face gently convex transversely, curving slightly longitudinally. Dorsal face strongly convex, and showing a slight tendency to become angular at the center, a little concave longitudinally. From the direc tion of the surface lines the aperture appears to have been oblique. Operculum unknown. Surface marked by lines of growth that on the dorsal side are nearly tran-verse, and on the ventral side arched forward; traces of fine longitudinal lines are shown in the matrix of the ventral side.

Dimensions.--Length, about 90mm; breadth of aperture, 23mm; height of aperture, about 14mm. The apex is broken away, and the aperture a little crushed by compression.

The most nearly related American species is Hyolithes excellens, Billings (Pal. Foss., vol. ii, pt. 1, p. 70, fig. 39, 1874), from Smith's Sound,

Trinity Bay, New Foundland. Mr. Billings's description and figures lead me to think that the two species are closely allied, but still distinct, species.

The specific name is given in honor of the discoverer of the specimen illustrated, Mr. N. S. Shaler, Paleontologist of Harvard University. Formation and locality.-Lower Cambrian. Braintree argillites, Hayward's Quarry, South Braintree, Mass.

Genus PARADOXIDES, Brongniart.

PARADOXIDES HARLANI, Green.

Plate vii, fig. 3; plate viii, figs. 1, la-e; plate ix, fig. 1.

Paradoxides Harlani, Green, 1834, Amer. Journ. Sci., vol. xxv, p. 336. Rogers, Wm. B., 1856. Proc. Boston Soc. Nat. Hist., vol. vi, pp. 27–29, pp. 40-44; ibid., Stodder, p. 369. Rogers, Wm. B., also in Amer. Jour. Sci., 2d ser., vol. xxii, p. 296. Rogers, H. D., 1858. Geol. Penn., vol. ii, p. 816. fig. 590.

Ordway, 1861. Proc. Boston Soc. Nat. Hist., vol. viii, pp. 1–5; ibid., Jackson, p. 58.

Dana, 1863. Manual of Geology, p. 189, fig. 245. This figure appears in each subsequent edition of the Manual.

Compare Paradoxides spinosus, Boeck.

Mr. Green's original description is as follows:

Description." The contour of the buckler in this species cannot be satisfactorily determined from our present specimen; the anterior and posterior parts of it are well defined, but the cheeks on each side are either mutilated or obscured. The front is very much elevated above the surface of the cheeks. It rises a little before the anterior edge of the buckler, is rounded in front, and gradually tapers towards the middle lobe of the abdomen, with which it forms a regular continuation. On its posterior surface there are three transverse furrows; the upper one crosses it a little obliquely, and there is on each side above a considerable protuberance. The cheeks were, no doubt, in the form of spherical triangles, but whether the outer angles terminated in acute prolongations cannot, from our specimen, be determined. The organs of vision appear to be entirely wanting. There are two shallow depres sions on each side of the cheeks, commencing near the protuberances on the front, and running towards the lateral edges of the buckler. The posterior border of the buckler, where it joins the lobes of the abdomen, is marked by a transverse groove, nearly continuous with the lower transverse furrow on the front; this groove at its commencement appears to bifurcate outwards. The abdomen and tail cannot be distin guished from each other. There are seventeen distinct articulations in both. The middle lobe is very convex, and is separated from the lateral ones by a deep channel; it gradually tapers to an obtuse tip. In our specimen there is a small part of the tail of another trilobite deposited in this place, which at first sight appears to be a dislocated fragment

!

of our animal. The lateral lobes are flattened; the costal arches are very distinct near their insertion, and for about half their length, but towards their free extremities they are a good deal obliterated. There appears to have been a delicate membranaceous prolongation for a considerable distance beyond the solid portion of each rib. This organization is very apparent on the costal arches of the tail. There is a deep groove running obliquely over the upper surface of each rib. Length of the fossil about 9 inches; breadth, about 4 inches."

Mr. Green did not know the true locality of the specimen sent to him by Mr. Harlan, and it was not until twenty-two years after that Mr. W. B. Rogers announced the discovery of specimens of the same species at South Braintree, near Boston, Mass., identifying the locality of the specimen used by Mr. Green in his original description. The description is unaccompanied by figures, but fortunately Mr. Green made numerous casts of the type, one of which is now before me. It is the narrow form of the species, measuring 22cm in length by about 14cm in width across the back of the head, and 12cm across the widest portion of the thorax. The palpebral lobes and movable cheeks are broken away, also the posterior segment of the thorax and the pygidium is displaced. Mr. Green describes the species as having 17 thoracic segments; but in a very fine specimen now in the collection of the Boston Society of Natural History, 18 segments are shown between the head and pygidium; and Mr. Henry D. Rogers gives a very perfect figure with 18 thoracic segments.

Mr. Ordway, in making a comparison between that species and the Bohemian P. spinosus (Proc. Boston Soc. Nat. Hist., vol. viii, p. 3), gives an outline figure of the head of P. Harlani, which is evidently a restored figure made up from fragments.

ns.

At the request of Mr. J. D. Dana, Mr. F. B. Meek drew a figure of P. Harlani for the Manual of Geology, from more or less fragmentary specimens in Mr. Dana's collection. This is one of the best, but not the best (see Rogers's figure), representations of the species yet published; but in the presence of 19 segments in the thorax, and the short extension of the posterior pleuræ and other details, it varies from specimens before There is considerable variation in the species in the relative length and breadth of individuals. In a form similar to the type, the length is 21cm, and the greatest breadth of the thorax 10cm. In two broad specimens the length is 25cm and 35cm; the breadth of the thorax 16cm and 20cm., respectively. This variation is also shown in the pygidium, as may be seen by comparing figs. 16, c, d of plate viii. In the head the greatest variation is seen in the contour of the frontal margin and the gradual development of the frontal limb and rim. On the small specimens the frontal limb is very short and more or less rounded. With the increase in size, the space between the glabella and the marginal rim increases in width, and the latter broadens and flattens out. Our information respecting the postero-lateral spines of the head is limited. On the narrow form, fig. 1, plate ix, they extend back to a point opposite the

fourteenth thoracic segment; and the movable cheek, fig. 3, plate vii, shows a long, well-developed spine. There is a limited range of varia tion in the extension of the pleure of the thoracic segments, but the material for study is too limited to say what value may be placed upon it. In reviewing all the variations, I do not think that more than one species is indicated. A narrow and broad variety might be designated if thought desirable.

Of American species of the genus Paradoxides, P. Bennetti, Salter (Quart. Journ. Geol. Soc., vol. xv, p. 552, fig. 1, 1859), from Newfoundland, is the most nearly related. The figure accompanying Mr. Salter's description appears to have been taken from a distorted specimen, as the two specimens now before me, although imperfect, show a form very similar to that of P. Harlani.

Mr. Ordway has described the differences between P. Harlani, Green, and P. spinosus, Boeck (Proc. Boston Soc. Nat. Hist., vol. viii, pp. 1-5, 1861), and froin my own observations and comparisons I cannot but agree with Mr. Ordway that the two species are represented. Mr. Barrande considered P. spinosus and P. Harlani as one species (Bull. Geol. Soc. France, vol. xvii, pp. 545-547, 1860). Mr. Barrande had the cast of the imperfect specimen described by Mr. Green to compare with specimens of P. spinosus, and photographs of three specimens sent to him by Mr. W. B. Rogers. The two species are, however, very closely related. Formation and locality.-Lower Cambrian. Braintree argillites. Hayward's quarry, South Braintree, Mass.

Genus PTYCHOPARIA, Corda.

PTYCHOPARIA ROGERSI, n. sp.

Plate vii, fig. 2.

This species is known only by the central portions of the head, two specimens showing portions of the thorax.

and

Glabella cylindro-conical, rounding rather abruptly in front, posterior pair of glabellar furrows very faintly shown in one specimen; dorsal furrow strongly defined; occipital furrow rounded, well marked and extending out across the fixed cheeks; occipital ring rather narrow, rising at the center and extending backwards in a short, strong spine; fixed cheeks of medium width, moderately convex, and sloping forward to unite with the frontal limb; ocular ridges shown only on one specimen; starting a little back of the anterior end of the glabella, they extend obliquely backward to the small palpebral lobe; frontal limb rather narrow; it curves downward for a short distance in front of the glabella and then up to the frontal rim. The facial sutures cut the anterior margin so as to leave a narrow frontal limb, and then extend obliquely outward and backward to the palpebral lobe; back of this they extend obliquely outward to the posterior margin of the head.