386 H. Woodward — On a Neic Fossil Arachnide. 



whicli is rounded near its extremity, is only lialf the breadtli of the 

 most anterior one. On reaching the last segment, the two nearly 

 parallel longitudinal lines turn slightly outwards, and terminate on 

 its lateral border. The last segment bears the efferent aperture on 

 its under surfaces. No ornamentation is visible on any part of the 

 specimen. 



I was at first inclined to believe this singular Palseozoic fossil to 

 be quite new and undescribed ; but on turning to Mr. S. H. Scudder's 

 "Supplement to the Description of Articulates, being a Description of 

 Fossil Insects found on Mazon Creek, and near Morris, Grundy Co., 

 Illinois," contained in Mr. A. H. Worthen's Eeport on the G-eological 

 Survey of Illinois, vol. iii. (Geology and Palseontology), 1868, I 

 found, on p. 568, an obscure figure of an Arachnide, from the Coal of 

 Illinois (reproduced on our Plate IX.), which I believe to be generi- 

 cally identical with the Lancashire specimen above described. 



I subjoin Mr. Scudder's remarks thereon : — " The last of the 

 nodule-specimens (see PI. IX.', Fig. 2) is perhaps the most interest- 

 ing. I believe the remains to be those of an Arachnide. This is the 

 first discovery of a fossil spider in America, and, as far as I know, 

 only the fourth instance of the occurrence of Arachnidce in Carbon- 

 iferous strata. In 1835 and 1839, Corda first figured two species — 

 one a true Scorpion, the other a gigantic Pseudo-scorpion. Eecently, 

 Eoemer has described a true Spider, under the name of Protolycosa. 

 Thus, the three known Carboniferous Arachnidce represent three 

 distinct families of Octopods. The one figured here (Architarbus 

 rotundatus) seems to belong to a fourth family, being allied to the 

 PTialangidcB and to the Phrynidce. In its fragmentary state, one can 

 scarcely judge with certainty of its exact relationship. The arrange- 

 ment of the legs accords equally well with both families. The broad 

 attachment of the thorax to the abdomen is a phalangidan character- 

 istic, while the size and shape of the abdomen, the number of the 

 abdominal segments, and the crowded state of the central portions 

 of the basal ones, indicate closer affinities to the Phrynidm. 



" The under surfaces of the thorax and abdomen are exposed to 

 view, together with a fragment of one of the legs. The thorax is 

 nearly circular ; the arrangement of the coxge uniformly radiate ; the 

 two joints of one leg are of equal length, and broader at the apex 

 than at the base. 



"The abdomen is nearly as broad at the base as the thorax; it 

 broadens close to the base, and beyond is ovate. The first abdominal 

 segment is scarcely perceptible at the sides; very large in the 

 middle, crowding downwards the four succeeding segments, which 

 are short and bowed. The terminal three segments are long and 

 straight, the last having just at the tip, but on the under surface, 

 the circular anal opening. Laterally all the segments are deiiressed, 

 and thus a broad flat border is formed on either side. Length of 

 specimen, 18^ mill. ; length of thorax,. 8^ mill. ; breadth of the 

 thorax, 9 mill. ; greatest breadth of abdomen, 9 mill. ; length of 

 the longest abdominal segment, 1^ mill. ; length of one leg-joint, 

 3i mill. ; breadth of the flattened margin of abdomen, f mill." 



