Do not circulate HARVARD UNIVERSITY. LIBRARY OF THE MUSEUM or OOMPAEATIVE ZOOLOGY. GIFT OP ^JUjWto>^ ^3 >y^ -^^lcilaJUa 2l,SL,V^V^ ^ I g J^ 7 BULLETIN OF THE MUSEUM OF COMPAKATIVE ZOOLOGY AT HARVARD COLLEGE, IN CAMBRIDGE VOL. LVIII. CAMBRIDGE, MASS., U. S. A. 1913-1914. The Cosmos Press: E. W. Wheeler, Cambridge, U. S. A. CONTENTS. Page No. 1. — Notes on a collection of birds from the Sudan. Bj^ John C. Phillips. December, 1913 1 No. 2. — Explorations in the Gulf of Maine, Juh- and August, 1912, by the U. S. fisheries schooner Grampus. Oceanography and notes on the plankton. By Henry B. Bigelow. (9 plates). February, 1914 29 No. 3. — The Stanford expedition to Brazil, 1911, John C. Branner, Dii-ector. The Chilopoda of Brazil. By Ralph V, Ch.'VMBERLIN. (Opiates). April, 1914 149 No. 4. — Notes on the ontogeny of Paradoxides, with the description of a new species from Braintree, Mass. By Percy E. Raymond. (1 plate). April, 1914 223 No. o. — Notes on the ontogeny of Isotelus gigas Dekay. By Percy E. Raytvignd. (3 plates). April, 1914 245 No. 6. — Notes on a collection of birds from Yunnan. By Outram Bangs and John C. Phillips. April, 1914 265 No. 7. — Mammals from the Blue Nile valley. By Glover M. Allen. July, 1914 303 No. 8.— Reports on the scientific results of the expedition to the Tropi- cal Pacific in charge of Alexander Agassiz, on the U. S. fish commission steamer ''Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., commanding. XVII. Reports on the scientific results of the expedition to the Eastern Tropical Pacific in charge of Alexander Agassiz, by the U. S. fi.sh commission steamer "Albatross," from October, 1904, to [March, 1905, Lieut. Commander L. M. Garrett, U. S. N., commanding. XXVIII. Isopoda. By Harriet Richard.sonSe.\rle. August, 1914 359 No. 9. — A new Peripatus from Colombia. By Charles T. Brues. (2 plates). September, 1914 373 No. 10. — Oceanogi-aphy and plankton of Massachusetts Bay and adja- cent waters, November, 1912-May, 1913. By Henry B. Bigelow. (1 plate). November, 1914 383 No. 11. — New Miocene Coleoptera from Florissant. By H. F. Wick- ham. (16 plates). December, 1914 421 / >^ vf. J v% A ' ' I' ( ''III Bulletin of the Museum of ConipaWMt^;^,o£I^ffy, AT HARVARD COLL£jGE. Vol. LVIII. No. 1. NOTES ON A COLLECTION OF BIRDS FROM THE SUDAN. By John C. Phillips. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. December, 1913. Reports 01^1 the Scientific Results of the Expedition to the East- ern Tropical Pac;if ic, in charge of Alexander Agassiz, by the V. B. Fish Commission Steamer "Albatross," from October, 1904, to Marohi 5^v>05, Li£utenant Commander L. M. Garrett, U. S. N., Commanding, published or in preparation: — A. A. A. H. H. H. R. O. S. w, L. W. C. «. H. H. ■S. w w c. ■c. XVI. i» The Medusae. XXIII." The Sipho- AGASSIZ. V.B General Report on the Expedition. AGASSIZ. I.' Tliree Letters to Geo. M. Bowers, U. S. Fish Com. AGASSIZ and H. L. CLARK. The Echini. B.BIGELOW, B. BIGELOW. nophores. B.BIGELOW. XXVI.2' TheCteno- phores. P. BIGELOW. The Stomatopods. CARLGREN. The Actinaria. P.CLARKE. VI11.8 TheHydroids. . R. COE. The Nemerteans. J. COLE. XIX.19 The Pycnogonida. . H. DALL. XIV.i* The Molluslis. R. EASTMAN. VII.' The Sharlis' Teeth. GARMAN. XII." The Reptiles. J. HANSEN. The Cirripeds. XXVJl." The Schl- J. HANSEN zopods. HENSHAW. E. HOYLE. The Insects. The Cephalopods. C. KENDALL and L. RADCLIPFE. XXV. 2' The Fishes. A, KOFOID. III.3 IX.» XX.20 The Protozoa. A. KOFOID and J. R. MICHENER. XXII." The Protozoa. C. A. KOFOID and E. J. RIGDEN. XXIV.2' The Protozoa. P. KRUMBACH. The Sagittae. R. VONLENDENFELD. XXI." The Siliceous Sponges. H. LUDWIG. The Holothurians. H. LUDWIG. The Starfishes. H. LUDWIG. The Ophiurans. G. W. MiJLLER. The Ostracods. JOHN MURRAY and G. V. LEE. XVII." The Bottom Specimens. MARY J. RATHBUN. X."" The Crus- tacea Decapoda. HARRIET RICHARDSON. II.« The Isopods. W. E. RITTER. IV.< The Tunicates. ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. F. E. SCHULZE. XI." The Xenophyo- phoras. H. R. SIMROTH. The Pteropods and Heteropods. E. C. STARKS. XIII." Atelaxla. TH. STUDER. The Alcyonaria. JH. THIELE. XV.i« Bathysciadium. T. W. VAUGHAN. VI.« The Corals. R.WOLTERECK. XVIII." TheAm- phipods. R. V. CHAMBERLIN. The Annelids. » Bull. M. C. Z., Vol. XLVL, No. 4. April, 1905, 22 pp. 2 Bull. M. C. Z.. Vol. XLVL, No. 6, July, 1905, 4 pp., 1 pi. ^ Bull. M. C. Z . Vol. XLVL, No. 9, September, 1905, 5 pp.. 1 pi. * Bull. M. C. Z., Vol. XLVL, No. 13, January, 1906, 22 pp., 3 pis. . «j. Zi.. vol. ALiVi., iNO. o, jiuy, lyua, t pp., 1 pi. ^ Bull. M. C. Z . Vol. XLVL, No. 9, September, 1905, 5 pp.. 1 pi. * Bull. M. C. Z., Vol. XLVL, No. 13, January, 1906, 22 pp., 3 pis, « Mem. M. C. Z., Vol. XXXIIL, January, 1906, 90 pp., 96 pis. e Bull. M. C. Z., Vol. L., No. 3, August, 1906, 14 pp.. 10 pis. 1 Bull. M. C. Z.. Vol. L., No. 4, November, 1906, 26 pp., 4 pis. « Mem. M. C. Z., Vol. XXXV., No. 1, February, 1907, 20 pp., 15 » Bull M. C. Z., Vol. L., No. 6, February, 1907, 48 pp., 18 pis. w Mem. M. C Z., Vol. XXXV, No. 2, August, 1907, 56 pp., 9 pis, » Bull. M. C. Z., Vol. LI., No. 6. November, 1907, 22 pp., 1 pi, Z., Vol. LIL, No. 1, June, 1908, 14 pp., 1 pi. , Vol. LI., No. 6. November, 1907, 22 pp., . •2 Bull. M. C. Z., Vol. LIL, No. 1, June, 1908, 14 pp., 1 pi. " Bull. M. C. Z., Vol. LIL, No. 2, July, 1908, 8 pp., 5 pis. " Bull. M. C. Z., Vol. XLIIL. No. 6, October, 1908, 285 pp., 22 pis. "Bull. M. C. Z., Vol LIL, No. 5, October, 1908, 11 pp., 2 pis. 18 Mem. M. C. Z., Vol. XXXVII., February, 1909, 243 pp., 48 pis. ^ rr Tr„i XXXVIII., No. 1, June, 1909, 172 pp., 5 pis.. 3 maps. ^11. , No. 9, June, 1909, 26 pp., 8 pis. iVieill. IVJL. Kj, /j., V ui. .^v..ri-.^i. V XX., x- cux ixo-i jf , xcrui?, ^- " Mem. M. C. Z.. Vol. XXXVIII., No. 1, June, 190! i« Bull. M. C. Z., Vol. LIL, No. 9, June, 1909, 26 pp., 8 pis. " BuU. M. C. Z., Vol. LIL, No. 11, August, 1909, 10 pp., 3 pis. ^oBuU. M. C. Z.. Vol. LIL, No. 13, September, 1909, 48 pp., 4 pis. ii Mem. M. C. Z., Vol. XLL, August, September, 1910, 323 pp.. 56 pis. 22 Bull. M. C. Z., Vol. LIV., No. 7, August, 1911, 38 pp. :. C. Z., Vol. XXXVIII.. No. 2, December, 1911, 232 pp., 32 pla. n 7, Vol. T.TV . No. 10. Februarv. 1912. 16 PP.. 2 pis ■22 BUU. JVX. *J. //., vol. i-ii.\ ., INO. /, AUgUSl), i.»li, OO pp. « Mem. M. C. Z., Vol. XXXVIII.. No. 2, December. 1911. 232 pp ^4 Bull. M. C. Z., Vol. LIV., No. 10, February, 1912, 16 pp., 2 pis 2» Mem. M. C. Z., Vol. XXXV., No. 3, April, 1912, 98 pp., 8 pis. 2« Bull. M. C. Z.. Vol. LIV., No. 12, April, 1912, 38 pp., 2 pis. «' Mem M. C. Z., Vol. XXXV, No. 4, July, 1912, 124 pp.. 12 pis. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE. Vol. LVIII. No. 1. NOTES ON A COLLECTION OF BIRDS FROM THE SUDAN. By John C. Phillips. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. December, 1913. No. 1. — Notes on a collection of birds from the Sudan. By John C. Phillips. The following collection was obtained by Dr. G. M. Allen and the writer on the Blue-Nile and Binder Rivers, in Sennar, Sudan, from December 25, 1912 to February 25, 1913. A few birds were also collected at Luxor and Cairo on the way up and down the Nile. A visit was made to the mountain at Fazogli, a place which Mr. A. L. Butler tells me he, found very rich in birds in May. Our visit in January yielded very little indeed, and this suggests considerable local seasonal movements of resident species, which is borne out by the various excellent papers of Butler on Sudan birds (Ibis, 1905, 1908, and 1909). Since the Sudanese birds have been rather carefully studied in recent years, by Reichenow, Erlanger, Butler, Ogilvie- Grant, and others, it is not worth while to append many notes. I have therefore confined myself mostly to the status of some of the species as we found them in the winter months. Many of the winter birds of this region are European migrants, some are visitors from Egypt, while a large number are resident spe- cies and referable mostly to Abyssinian types. Thus the avifauna of the upper Blue-Nile is quite different from that of the White-Nile on account of this Abyssinian element. There is no true desert near the Blue-Nile and Dinder Rivers, the soil being a deep loess deposit, the so-called "cotton-soil," which in the dry season becomes baked and cracked into great cakes. These contraction-cracks make travel very unconfortable. The entire region is wooded with a widely scat- tered thorn-forest, nearly leafless by mid-winter. The ground is covered by high grass which is largely burnt off in December and January. Along the banks of the river the scenery is a little more diversified. Palms, fig trees, Adansonias, and vines form thickets in which many species hide while going to water. A few isolated rocky hills or gebels protrude abruptly from the plain. P^rom Gebel Fazogli eastward they begin to form the foothills of Abyssinia. The larger birds we did not have time to collect or preserve to any extent. Among the more striking may be mentioned the enormous numbers of European cranes present on the lower Blue-Nile, and also the Crowned cranes in much smaller numbers. Anatiflae are scarce on the Blue-Nile, on account of its sandy character, the Egyptian 4 bulletin: museum of comparative zoology. goose and the Comb duck being the commonest species. We saw the Secretary bird once only. It is rare in the Sudan. The most abundant scavenger is the Egyptian kite which is bold past all belief, while there are several species of vultures. The Bate- leur eagle and the Great river eagle, HaUaetus vocifcrus are often seen. We saw the Sacred ibis on the Blue-Nile in great flocks, the Glossy ibis much less often, while pelicans of probably two species were noted occasionally. The huge Scavenger stork, Leptoptilus crumeniferus was nearly everx^vhere a constant camp attendant. The Common stork and the Saddle-back stork were seen here and there. Little egrets, Bubulcus ibis followed the game, especially Buffalo, in large flocks, while a large White heron Herodias alba, or Mcsophoyx brachyrhyncha was seen only once. Much remains to be worked out concerning the seasonal movements of resident birds. The sharply contrasted wet and dry season is very marked in this part of the Sudan. On the Abyssinian border showers occasionally occur in winter, but this never happens farther westward. Our route was from Sennar to Fazogli on the north bank of the Blue-Xile, then back to Abu Tiga by the same road and across to the Dinder River which is only two days journey from the Blue-Nile at this point. On the Dinder we travelled S. E. to Um Orug Isle and then back to Sennar by forced marches. We did little bird collecting on the Dinder as our time there was short. We used camel transport for the entire journey. To Dr. Allen belongs the credit of most of the actual collecting and I have to thank iSIr. Outram Bangs for most valuable aid with identifi- cations. Phasianidae. Francolinus clappertont Children. 5 specimens; 3 d^'s, Bados, Blue-Nile, 7 Jan.; El Garef, 8 Jan.; Magangani, 1 Feb. Very plentiful in tall grass near the river. Ptilopachys fuscus fuscus (Vieillot). One cf, Fazogli, Blue-Nile, 18 Jan. Only seen on Gebel Fazogli. PHILLIPS: BIRDS FROM THE SUDAN. O NUMIDIDAE. NuMiDA PTiLORHYNCHA PTILORYNCHA (Lichtenstein). 2 9 's, Gabardi, Blue-Xile, 31 Dec; El Garef, 9 Jan. Seen in enormous numbers along our whole route. Pteroclididae. Pteroclis quadricinctus Temminck. Pair, Fazogli, Blue-Nile, 18 Jan, and 20 Jan. Common along Blue-Nile. Treroxidae. ViNAGO WAALiA WAALiA (Gmelin). 3 specimens; 1 cf, Gebel Fazogli, Blue-Nile, 18 Jan.; Um Orug, Binder, 16 Feb. As has often been remarked, this bird is almost confined to the fig tree. We did not find it in large numbers. COLUMBIDAE. COLIIMBA GUINEA PHAEONOTA (G. R. Gray). 1 9 , Roseires, Blue-Nile, 13 Jan. Not a common bird. Seen in pairs here and there in rocky places. Peristeridae. TuRTUR iSABELLiNUS Bonaparte. 1 cf , Sennar, Blue-Nile, 25 Dec. This can not be a common bird for our region, although we may have seen it and mistaken it for Strcptopclia vinacea schoanus. Streptopelia decipiens (Finsch & Hartlaub). 1 cf , El Serifa, Blue-Nile, 15 Jan. Very common. b bulletin: museum of comparath'e zoology. Streptopelia vixacea schoaxus (Xeumaiin). 3 cf's, Karkoj, Blue-Xile, 30 Dec. A very abundant bird. Stigmatopelia senegalexsis aequatorlvlis (Erlanger). 1 d", Karkoj, 30 Dec. Not so common as some other doves. Oexa capexsis (Linne). Pair, Sennar, Blue-Nile, 25 Dec; Aradeiba, 22 Jan. Very abundant, very tame and perfectly noiseless; in flocks. Chalcopelia afra Linne. 3 specimens; 2 d"s, Bados, Blue-Xile, 2 Feb.; Singa, 2S Dec; Magangani, 28 Jan. Abundant; in pairs or singly. Cilvradriidae. HoPLOPTERUS spixosus (Linne). 1 9 , Roseires, Blue-Xile, 13 Jan. A very common species. Stephaxibyx melaxopterus (Cretzschmar), 1 9 , Sennar, Blue-Nile, 27 Dec. Aegialitis dubia (Scopoli). Pair, Luxor, Egypt, 3 March; Roseires, Blue-Xile, 13 Jan. Common migrant. HiNLiNTOPUS HiMAXTOPUS (Linne). 1 9, Roseires, 11 Jan. Common in suitable spots along Blue-Xile. PHILLIPS: BIRDS FROM THE SUDAN. 7 AcTiTis HYPOLEUCUS (Linne). 3 specimens; 2 cf's, Roseires, Blue-Nile, 24 Jan.; Magangani, 27 Jan. Common. Tringa nebularia (Gunnerus). 1 cf, Roseires, Blue-Nile, 24 Jan. Common migrant. PisoBiA MiNUTA (Leisler). 1 9 , Galegu, Binder, 19 Feb. We found it rare, but Butler states that it is common. ROSTRATULA BENGALENSIS (Linne). Pair, Abiad, Binder, 14 Feb. Not common. CURSORIIDAE. Pluvianus aegypticus (Linne). 1 cf , Singa, Blue-Nile, 28 Bee. Very abundant. Oedicnemidae. BuRHiNUS SENEGALENSis Swainson. 1 9 , Magangani, 30 Jan. Very common on Blue-Nile and Binder. Otididae. EupoDOTis ARABS (Linne). 1 adult 9 , Gabardi, Blue-Nile, 30 Bee. Uncommon. x\rdeidae. Ardea melanocephala Vigors & Children. 1 9 , Abu Zor, Blue-Nile, 5 Jan. Common. 8 bulletin: museum of comparative zoology. BuBULCUS IBIS (Linne). 1? sex, Aradeiba, Blue-Nile, 15 Jan. Common. The stomach of this specimen contained two lizards, two locusts, and a butterfly. Falconidae. Circus macrurus (S. G. Gmelin). Pair, ]Magangani, Blue-Nile, 28 Jan.; Ereifa el Dik, Binder, 10 Feb. Common about marshy spots. Melierax met abates (Heuglin). 2 9 's, Ereifa el Dik, Binder, 11 Feb.; Singa, Blue-Nile, 29 Dec. Very common everywhere. The commonest hawk. Microxisus gabar (Baudin). 3 9 's, Galegu, Binder, 20 Feb.; ]\Iagan^ani, Blue-Nile, 28 Jan. Not common. MiCRONisus NIGER (Vieillot). Pair, Roseires, Blue-Nile, 24 Jan.; Gabardi, 31 Bee. Uncommon. AsTUR SPHENURUS (Riippell). 1 d^ , Roseires, Blue-Nile, 23 Jan. Only one seen. LoPHOAETUS OCCIPITALIS (Baudin). 1 9 , Roseires, Blue-Nile, 11 Jan. Seen only in a few places, uncommon. TCauptfat.co monogrammicus monogrammicus (Temminck). 1 cf , Fazogli, Blue-Nile, 16 Jan. Uncommon. Shot with a grass-rat (Arvicanthis) in the claws. PHILLIPS: BIRDS FROM THE SUDAN. 9 Falco biarmicus tanypterus (Schlegel). 1 d", Singa, Blue-Nile, 28 Dec. Not^^common. Falco ruficollis Swainson. Pair, El Garef, Blue-Nile, 8 Jan.; Magangani, 25 Jan. Not common. Cerchneis tinnunculus tinnunculus (Linne). 1 cf and 3 9's; Lakandi, Blue-Nile, 6 Feb.; Karkoj, 30 Dec; Roseires, 25 Jan. ; Sennar, 25 Dec. Common everywhere as a migrant. Stridgidae. Bubo maculosus cinerascens Guerin. 2 cf 's, Magangani, Blue-Nile, 27 Jan.; Abu Tiga, 6 Feb. Seen and heard in several places where trees were large. Otus capensis capensis (Smith). 3 d^'s, Abu Zor, Blue-Nile, 21 Jan.; El Mesharat, 5 Feb.; El Serif a, 4 Jan. Common every^vhere in thorn forest. Athene noctua glaux (Savigny). 1 cf , Cairo, Egypt, 14 Dec. Psittacidae. Poicephalus meyeri meyeri (Cretzschmar). Pair, El Serif a, Blue-Nile, 14 Jan.; Abu Zor, 6 Jan. Common on the Nile about Roseires but not seen north of El Mesharat. 10 bulletin: museum of comparative zoology. Palaeoenis torquatus docilis (Vieillot). 2 cf 's, Singa, Blue-Nile, 28 Dec; El Garef, 9 Jan. Very common over the region covered. Breeding freely in Jan. and Feb. in holes high up in trees. CORACIIDAE. CoRACiAS ABYSSiNicus Boddaert. 3 specimens, 2 cf's and 1 9 , El Garef, Blue-Nile, 9 Jan.; El Sa- bonabi, 2 Jan.; El Serif a, 22 Jan. Very common. The showiest bird of this region during the winter. CoRACius NAEVius NAEVius Daudin. 1 cf, Aradeiba, Blue-Nile, 21 Jan. Rare. Only two seen. Alcedinidae. Ceryle rudis rudis (Linne). 1 cf , Roseires, Blue-Nile, 13 Jan. Common along the river. CORYTHORNIS CYANOSTIGMA (Ruppell). 2 specimens, 1 d^, Abiad, Dinder, 14 Feb.; 1, Roseires, Blue-Nile, 13 Jan. Not common. In pairs along the river and the pools of the Dinder. Halcyon chelicuti (Stanley). 1 9 , El Mesharat, Blue-Nile, 2 Jan. Uncommon and found in the thorn forest at some distance from water. Bucerotidae. LoPHOCEROS HEMPRiCHi Ehrenberg. 1 9 , El Mesharat, Blue-Nile, 2 Jan. Very common all through the thorn bush in flocks of three or four. PHILLIPS: BIRDS FROM THE SUDAN. 11 Very tame. Sits low in the trees and the extraordinarily soft " Weet, weet," grows louder and louder until reaching a crescendo which is accompanied by an upright position and flapping of wings. This is almost the only bird sound heard at the heat of the day. LoPHOCEROS NASUTUS NASUTUS (Linne)- 1 cf , ^Slagangani, Blue-Nile, 29 Jan. Common in certain places only. Usually singly and in the tops of high trees. Very wild. A loud, clear, double note with opening and closing of wings. Upupidae. Upupa epops epops Linne. Pair, El Serifa, Blue-Nile, 14 Jan.; El Garef, 1 Feb. Note. Upupa butlcri of Madarasz (Ann. Mus. nat. Hung., 9, p. 339) appears to be based on a small example of the European Hoopoe, which is a common winter resident in the Sudan. Upupa epops major Brehm. 1 d^, Galegu, Binder, 20 Feb. Probably common. This specimen certainly belongs to the large- billed form. Irrisoridae. Irrisor erythrorhynchus niloticus Neumann. 1 cf, Fazogli, Blue-Nile, 16 Jan. Common in certain areas in large flocks. Rhinopomastus minor (Riippell). 2 cf' s and 1 9 , Singa, Blue-Nile, 28 Dec; Bados, 6 Jan. Fairly common, usually seen singly. Meropidae. Melittophagus pusillus ocularis Reichenow. 1 c^ and 2 9 's, Roseires, Blue-Nile, 13 Jan.; El Mesharat, 2 Jan. Common in pairs or broods. 12 bulletin: museum of comparative zoology. Melettophagus frenatus Hartlaub. 3 d^'s and 2 9 's, El Mesharat, Blue-Nile, 4 Jan.; El Sabonabi, 1 Jan. Common, roosting in holes in banks. MeROPS ORIENTALIS CLEOPATRA Nicoll. 1 cf and 3 9 's, Luxor, Eg;^T3t, 3 March; Fazogli, Blue-Nile, 19 Jan.; Magangani, 26 Jan.; x\bu Zor, 5 Feb. None of these show any variation; therefore the Blue-Nile birds are probably migrants from Egypt; seen only three times on the Blue- Nile would indicate this as its southern limit. Merops nubicus Gmelin. 3 d^'s and 2 9 's, Galegu, Binder, 12 Feb.; Abiad, 14 Feb. Very common on the Binder but rather rare on the Blue-Nile. Roosted in flocks in thick trees. Flies very high with habits of a swallow. Caprimulgidae. ScoTORNis climacurus (Vicillot). 6 cf's and 4 9 's, Magangani, Blue-Nile, 31 Jan., 26 Jan., 27 Jan., 28 Jan.; Singa, 27 Bee; Roseires, 13 Jan., 23 Jan.; Ereifa el Bik, Binder, 10 Feb. Very common about the high grass of both rivers. Sprung often in the woods during the day. When flying it makes a single clucking noise. These birds often eat their prey on the ground. Macrodipteryx macrodipteryx (Afzelius). 1 d^ and 2 9 's, Magangani, Blue-Nile, 30 Jan., 27 Jan.; El Serifa, 19 Jan. We found this remarkable bird rare and local. We did not see more than three or four males. They fed over high elephant grass and never crossed open spots. They were about early in the evening and each male seemed to have a separate and distinct range. PHILLIPS: BIRDS FROM THE SUDAN. 13 Caprimulgas aegyptius Lichtenstein. 3 cf 's and 3 9 's, Magangani, Blue-Nile, 25 Jan., 26 Jan., 27 Jan., 29 Jan., 31 Jan.; Roseires, 13 Jan. This may be C. a. saharae Erlanger, the paler resident African race, but no material is at hand for comparison. Common over the high grass at dusk. In large areas of elephant grass they were very numerous. Dr. Allen found ground crickets and cockroaches in the stomach of one of these birds which was shot early in the evening. Caprimulgus eleanorae Phillips. Proc. Biol. soc. Washington, 1913, 26, p. 167. Type, adult 9 ^I. C. Z. No. 63,436, taken at Fazogli, Blue-Nile, Sudan, 15 Jan., 1913. Description. — Most nearly like C. monticola of India, of which it appears to be the African representative. In general color very much like the gray examples of C. monticola, but at once distinguished by the spots on the three outer primaries, being small and round and confined wholly to the inner web, instead of being large and extending across both webs of 2d, 3d, and 4th primaries. In the new species the spot on the first primary is only 7 mm. in diameter, while on the third primary it is about 12 mm. Wing, 185 mm.; culmen to base of fore- head, 28 mm.; exposed culmen, 11 mm.; tarsus, 20 mm. The char- acters of the male are unknown. Remarks.— This species differs from all African species of somewhat similar general coloration by its much larger size, equal in fact to C. monticola. Only one specimen was taken. MiCROPODIDAE. Tachorxis PARVUS PARVUS (Lichtenstein). 1 d^, Roseires, Blue-Nile, 13 Jan. Common along rivers, especially about deleb palm trees. 14 bulletin: museum of comparative zoology. coliidae, Colius leucotis leucotis Riippell. 3 specimens, 2 cT's and 1 9, Magangani, Blue-Nile, 26 Jan.; El Mesharat, 4 Jan. Common. In small flocks. Colius macrurus (Linne). 1 9 , Gabardi, Blue-Nile, 31 Dec. We found it rare, but Butler gives it as common. Musophagidae. Chizaerhis zonura Riippell. 1 cf , Fazogli, Blue-Nile, 19 Jan. Only seen near Fazogli. CUCULIDAE. Chrysococcyx klaasi (Stephen). 3 specimens, 2 c/"s and 1 9 , Roseires, Blue-Nile, 25 Jan. Rare : only seen once in a large fig tree in the village. Centropus superciliosus Hemprich & Ehrenberg. 2 9 's, Singa, Blue-Nile, 28 Dec; El Garef, 9 Jan. Uncommon. Indicatoridae. Indicator indicator (Gmelin). 3 specimens, 1 cf and 2 9 's, Um Orug, Dinder, 17 Feb., 18 Feb. Seen mostly on the upper Dinder where honey was plentiful. Indicator minor diademata (Riippell). 1 9 , Um Orug, Dinder, 17 Feb. Probably not so common as I. indicator. phillips: birds from the sudan. 15 Capitonidae. Lybius tridactylus (Gmelin). Pair, Roseires, Blue-Nile, 25 Jan., 13 Jan. Uncommon; only seen once or twice. Lybius vieilloti (Leach). 2 cT's and 1 9 , Magangani, Blue-Nile, 30 Jan.; El Mesharat, 4 Jan. More common than L. tridactylus but only seen here and there. Barbatula chrysocoma chrysocoma (Temminck). 2 d^'s and 1 9 , El Garef, Blue-Nile, 9 Jan.; Magangani, 30 Jan. This may be zedlctzi of Neumann on geographic grounds, but does not seem to agree with his description. Ours is more like true chryso- coma. This bird was very common on the Binder. Nest in the dead limb of a tree, with full grown young on Jan. 8; food of berries was being brought to the young, and was ejected after being eaten. PiCIDAE. Dendromus nubicus nubicus (Gmelin). 3 specimens, 2 cT's and 1 9 , El Garef, Blue-Nile, 8 Jan., 10 Jan.; Fazogli, 20 Jan. Common. Has a cackling note which is heard commonly through the thorn forest. In pairs. Mesopicus goertan abessinicus (Reichenow). Pair, Galegu, Binder, 20 Feb. ; Magangani, Blue-Nile, 28 Jan. Rare, only seen two or three times. Bendropicus obsoletus obsoletus (Wagler). 4 cf 's and 1 9 , Mesharat Kuka, Binder, 9 Feb.; Bados, Blue-Nile, 2 Feb.; Wad Shara Shara, 8 Feb.; Abu Tiga, 7 Feb. Rather uncommon. 16 bulletin: museum of comparative zoology. HiRUNDINIDAE. RiPARiA MINOR (Cabanis). 2 cf 's and 1 9 , Fazogli, Blue-Nile, 20 Jan.; Serifa, 21 Jan. Breeding in holes in bank near Fazogli and Roseires towards the end of January. Numerous. Chelidon aethiopica Blanford. 2 d^'s, Abiad, Dinder, 14 Feb. Fairly common, the common resident swallow according to Butler. Chelidon griseopyga Sundeval. Pair, Fazogli, Blue-Nile, 20 Jan. We found a colony of these birds near Fazogli breeding in burrows dug in the hard clay on open, level ground; the burrows ran parallel to the surface. One that we dug out was ten feet long and had a depth of from six to ten inches. The nest was three feet from the end. The young were partly fledged on Jan. 25. Chelidon daurica rufula (Temminck). 1 d", Abiad, Dinder, 14 Feb. Rare. Muscicapidae. Melaenornis pammelaena Stanley. 1 cf , Roseires, Blue-Nile, 23 Jan. Probably a rare bird. Bradyornis pallidus granti Bannerman. Pair, Galegu, Dinder, 19 Feb.; Sabonabi, Blue-Nile, 6 Feb. Our bird is small and with underparts more strongly suffused with rufous, like B. granti of Bannerman, Bull. B. O. C, 1911, 27, p. 84. The differences do not seem to be more than subspecific. wing 85, tail feathers 71, tarsus 21, culmen 13. wing 80, tail feathers 69.5, tarsus 20, culmen 11.5. PHILLIPS: BIRDS FROM THE SUDAN. 17 The records from the Dinder River extend the range of this small form into the eastern Sudan. Uncommon. Batis orientalis orientalis (Heuglin). 1 cf and 2 9 's, Magangani, Blue-Nile, 26 Jan., 27 Jan. Fairly common. TcHiTREA viRiDis (Midler). 1 cT, Urn Orug, Dinder, IS Feb. Rare, only seen twice; in dense thickets. Pycnonotidae. Pycnonotus arsinoe (Hemprich & Ehrenberg). 2 d"s. El Mesharat, Blue-Nile, 4 Jan.; El Garef, 8 Jan. One of the commonest birds in thickets along the river. Timeliidae. Crateropus leucocephalus (Cretzschmar). 3 cf's, Abu Zor, Blue-Nile, 5 Jan; El Mesharat, 4 Jan. Fairly common in small flocks. Turdidae. Planesticus pelios pelios (Bonaparte). 1 cf and 2 9 's, El Garef, Blue-Nile, 8 Jan.; 18 Jan.; Fazogli, 8 Jan. Not common ; seen singly in the southern part of our region. Phoenicurus phoenicurus phoenicurus Linne. 1 cf and 2 9 's, Magangani, Blue-Nile, 29 Jan., 1 Feb.; El Serifa, 14 Jan. Not common. 18 bulletin: museum of comparative zoology. Cyanosylvia svecica volgae (Klemschmidt). 1 cf , Cairo, Egypt, 12 Dec. This may be C. svecica svecica in winter plumage. Saxicola torquata rubicola Linne. Pair, Cairo, Egypt, 12 Dec. Saxicola torquata maura (Pallas). 1 d', and 2 9 's, Abiad, Dinder, 14 Feb.; Urn Orug, 16 Feb.; Galegu, 20 Feb. Not common. In open places on grass. Oenanthe oenanthe oenanthe (Linne). 1 cf, Roseires, Blue-Nile, 25 Jan. Said to be a common species but we did not find it so. Oenanthe hispanica xanthomelaena Hemprich & Ehrenberg. 2 cf 's and 1 9, x\bu Tiga, Blue-Nile, 7 Feb.; Fazogli, 16 Jan.; Singa, 28 Dec. Common in open woods. Shy and seen singly. Oenanthe isabellina Cretzschmar. 1 cf , Sennar, Blue-Nile, 25 Dec. Common in durrah fields round Sennar, not seen far south. Oenanthe melanoleuca melanoleuca (Giildenstaedt). 1 d^ and 2 9 's, Mesharat Kuka, Dinder, 9 Feb. ; Singa, Blue- Nile, 28 Dec; Sabonabi, 6 Feb. Fairly common. phillips: birds from the sudak. 19 Sylviidae. Agrabates galactotes galactotes Temminck. 1 cf , Khamisa, Dinder, 23 Feb. This specimen is apparently an extreme of the European species, (not minor of Cabanis) and has a wing of 85 mm. with a broad sub- terminal tail band. According to Hartert this bird migrates into the Sahara or to south of it. Only seen once or twice. Cisticola semitorques semitorques Heuglin. Pair, Abiad, Dinder, 14 Feb.; Beit el Wahsh, Dinder, 13 Feb. Rather common. Cisticola ferrugixea (Heuglin). 1 9 , Fazogli, Blue-Nile, 20 Jan. Uncommon, in thick grass. Cisticola cisticola cisticola (Temminck). Pair, Cairo, Egypt, 12 Dec. Cisticola margixata (Heuglin). 2 9 's, Karkoj, Blue-Nile, 31 Dec; Wad Shara Shara, 8 Feb. Hypolais pallida pallida (Hemprich & Ehrenberg). 1 d^, Singa, Blue-Nile, 27 Dec. Phylascopus collybite collybite (Vieillot). 1 cf and 2 9 's, Cairo, Egypt, 12 Dec. ; Luxor, 3 jNIarch. Apalis pulchella (Cretzschm). 3 cf's, Magangani, Blue-Nile, 26 Jan., 27 Jan.; El Sabonabi, 1 Jan. Fairly common. 20 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. Sylviella beachyura nilotica Neumann. 2 d"s, Ereifa el Dik, Binder, 10 Feb.; Galegu, Binder, 20 Feb. Wings 53 and 59 mm. Rare. Eremomela elegans (Heuglin.) 1 d" and 2 9 's, Magangani, Blue-Nile, 30 Jan. These were the only ones seen. Camaroptera griseiviridis griseiviridis v. Miiller. 3 d^'s and 1 9 , Magangani, Blue-Nile, 28 Jan.; El Garef, 8 Jan. Common in dense thickets and on the ground. Singly. Prinia mystacea Riippell. 4 d^'s and 3 9 's, Magangani, Blue-Nile, 28 Jan., 30 Jan.; Singa, 27 Dec.; Fazogli, 20 Jan.; El Mesharat, 4 Jan.; El Sabonabi, 1 Jan.; Beit el Wahsh, Binder, 13 Feb. The common warbler of the country. Prinia gracilis deltae Reichenow. 3 9 's, Cairo, Egypt, 12 Dec. Prionopidae. Prionops poliocephala (Stanley). 1 d", El Mesharat, Blue-Nile, 4 Jan. Occasional, in small flocks. Laniidae. Lanius excubitor pallidirostris (Cassin). 1 9 , Sennar, Blue-Nile, 25 Dec. Only seen once. PHILLIPS: BIRDS FROM THE SXTDAN. 21 Lanius nubicus (Lichtenstein). 1 c^ and 2 9 's, Shiga, Blue-Nile, 27 Dec. Common on Blue-Nile. Laniarius erythrogaster (Cretzschmar) . 3 cf 's and 3 9 's, Magangani, Blue-Nile, 27 Jan., 31 Jan.; Bados, 7 Jan.; Abu Zor, 5 Feb. ; El Garef, 8 Jan. Common, in pairs in thickets. Shy. These birds have responsive notes. Dr. Allen followed this up and shot the bird of a pair having the first note. It was the male. These birds fed much on the ground and we frequently caught them in our traps. One of the most beautiful of the Blue-Nile birds. Dryascopus cinerascens Hartlaub. 2 d"s and 3 9 's, Magangani, Blue-Nile, 26 Jan., 30 Jan.; EI Mesharat, 4 Jan. ; El Sabonabi, 1 Jan. ; El Garef, 8 Jan. Fairly common; in trees. POMATORHYNCHUS BLANFORDI (Sharpc). 1 cf , Sennar, Blue-Nile, 25 Dec. Not common ; low in thickets, and on ground. PoMATORHYNCHUS REMiGiALis (Finsch & Hartlaub). Pair, Roseires, Blue-Nile, 12 Jan.; Magangani, 26 Jan. Uncommon. Butler found it far from water and regards it as com- mon. NiLAUS AFER AFER Latham. 1 d" and 2 9 's, Urn Orug, Dinder, 17 Feb.; Singa, Blue-Nile, 28 Dec. ; Fazogli, 19 Jan. Not very common. Paridae. MeLANIPARUS NIGER LEUCOMELAS Riippell. Pair, Gabardi, Blue-Nile, 31 Dec; Mesharat Kuka, Dinder, 9 Feb. Fairly common. 22 bulletin: museum of compakative zoology. Necterinidae. Hedydipna metallica (Lichtenstein). Pair, Galegu, Dinder, 20 Feb.; Durraba, Dinder, 22 Feb. Rare. Necterinia pulchella (Linne). 8 cf 's, Magangani, Blue-Nile, 27 Jan.; Abu Zor, 5 Feb.; El Mesh- arat, 2 Jan. ; Durraba, Dinder, 22 Feb. Common among flowering thorn bushes. These specimens are in all stages of immature plumage. Motacillidae. MoTACiLLA alba Linne. 5 cf's and 1 9 , Roseires, Blue-Nile, 13 Jan.; Cairo, Egypt, 12 Dec. Confined almost entirely to village streets in the southern part of the Blue-Nile, but farther north it is everywhere. A very common migrant. BuDYTES flava (Linne). ■ 3 d^'s, El Garef, Blue-Nile, 1 Feb., 2 Feb.; Abiad, Dinder, 14 Feb. In flocks along river. Occasional. BuDYTES MOTACILLA FELDEGGI (Michah). 1 cf , Wad Shara Shara, 8 Feb. The only one seen. Anthus campestris campestris (Linne). 1 cf , Abiad, Dinder, 19 Feb. Only seen at the above locality. Butler calls it fairly common. Alaudidae. Melanocorypha bimaculata (Menetries). 1 cf , Beida, Dinder, 8 Feb. One large flock seen. PHILLIPS: BIRDS FROM THE SUDAN. 23 Galerida CRIST ATA ALTiROSTRis (Rrehm). 1 c^, Luxor, Egypt, 3 March. Enormous numbers in Egypt. Pyrrhul.\uda leucotis leucotis (Stanley). Pair, Sennar, Blue-Nile, 25 Dec; Abiad, Dinder, 25 Dec. Not common. In pairs. Frixgillidae. Petroxia dextata (Sundeval). 4 cT's, Karkoj, Blue-Nile, 31 Dec; El Sabonabi, 1 Jan.; Ereif el Dik, Dinder, 11 Feb. The common sparrow of the thorn forest. Passer domesticus arboreus Bonaparte. 2 cf 's, Khartoum, 23 Dec, 24 Dec Seen only as far as Singa but not farther south. Passer domesticus chephrexi Phillips. Proc Biol. Soc Washington, 1913, 26, p. 167. Type, Adult cf ^I. C. Z. No. 63,594 from Gizeh near Cairo, Egypt, 12 December, 1912. Description. — Like P. d. indicus but cheeks and ear-coverts much darker (smoke gra}', Ridgway, 1912) instead of whitish. Size similar to P. indicKs. Adult female very similar to P. indicus, but cheeks darker and grayer. Remarks. — Hartert (Vogel der Palaarktischen fauna, 1, p. 151), did not name this lower Nile race, his material being insufficient, but in a footnote in the list of species to the first volume he refers the Egyp- tian bird to niloticus of Nicoll & Bonhote (Bull. B. O. C, 22, p. 101). P. d. niloticus is apparently a local desert race closely resembling P. d. arboreus, from Kliartoum, and not the typical sparrow of lower Egypt, which Nicoll and Bonhote still refer to P. (/. indicus. Pair, Cairo, Egypt, 12 Dec. 24 bulletin: museum of comparative zoology. Passer swainsoni (Riippell). Pair, Singa, Blue-Nile, 27 Dec; Aradeiba, 21 Jan. Common along Blue-Nile. Serinus icterus icterus (Vieillot). 1 9 , Fazogli, Blue-Nile, 16 Jan. Uncommon. Emberiza caesia Cretzschmar. 3 9 's, Roseires, Blue-Nile, 13 Jan.; Sennar, 25 Dec. A fairly common migrant. Ploceidae. Steganura paradisea (Linne). 2 d^'s, Roseires, Blue-Nile, 23 Jan.; Abu Zor, 6 Jan. Common near Singa and Sennar but rare farther south; if in winter plumage we may not have noticed it. Urobrachya phoexicea (Heuglin). 1 cf , Abiad, Dinder, 14 Feb. In flocks; with other weaver-finches. Ploceipasser superciliosus (Riippell). 1 &, Gebel Maba, Blue-Nile, 14 Jan. Only one seen. QUELEA SANGUINIROSTRIS AETHIOPICA (Sundcval). 4 cf's and 1 9 , Gabardi, Blue-Nile, 31 Dec; Bados, 4 Feb.; Singa, 28 Dec. Exists in unbelievable thousands and is a great menace to agri- culture. PHILLIPS: BIRDS PROM THE SUDAN. 25 Lagonostica sexegala (Linne). 3 cf 's and 2 9 's, El Garef, Blue-Nile, 8 Jan. These birds belong to some raee of the above species, but on geo- graphical grounds do not agree with Neumann's L. s. crythreae or aboyoms. It may be a new race. We did not obtain hrunneiceps, though it appears to be the common bird of the region. Common in little flocks. Pytilia citerior Strickland. 2 9 's, Magangani, Blue-Nile, 27 Jan. ; El Garef, 8 Jan. These specimens are in very young plumage. It is not possible to place them imder their proper subspecies from Zedlitz' review of the species, Ornith. monatsb., 18, p. 171. Uncommon. Hypochaera ultr.\marixa (Gmelin). 3 cf's and 2 9 's, Singa, Blue-Nile, 27 Dec; Regeba, 22 Jan.; Karkoj, 31 Dec. A very common species; in small flocks. AiDEMOSYNE CANTANS ORiEXTALis Loreuz & Hcllmayr. 3 cf' s, Durraba, Dinder, 22 Feb.; Sennar, Blue-Nile, 25 Dec. Occasional. Estrilda cixerea (Vieillot). Pair, El Garef, Blue-Nile, 10 Jan. Fairly common. Uraeginthus bengalus bengalus (Linne). 5 cf 's and 2 9 's, El Mesharat, Blue-Nile, 2 Jan.; Singa, 27 Dec; Magangani, 27 Jan. ; Beit el Wahsh, Dinder, 13 Feb. According to Reichenow, the above form ranges over the whole Sudan. The Sennar bird is therefore M. h. bcngaius and not M. h. shoenni or M. b. pcrpallidus of Neumann, Journ. f. ornith., 1905, p. 350. Very common, seen ever;y"where. 26 bulletin: museum of comparative zoology. Hyphantornis abyssinicus abyssinicus (Gmelin). 1 cf , El Mesharat, Blue-Nile, 4 Jan. Fairly common, in small flocks around durrah fields. Xanthophilus galbula Riippell. 1 cf and 2 9 's, Singa, Blue-Nile, 27 Dec. ; Magangani, 26 Jan. Common. This is the species which Butler (Ibis, 1905) describes as the commonest weaver, but the great flocks seemed to us to be mostly Quelea, the Red-billed weaver. Eulabetidae. Lamprotornis purpuropterus aeneocephalus Heuglin. 1 d^, Gabardi, Blue-Nile, 31 Dec. Tail, 8 inches, tail feathers, 7.3 inches. This is maximum accord- ing to the Catalogue of the birds in the British Museum, but less than Butler's maximum for Kordofan birds (Ibis, 1905, p. 324). Very common. Lamprocolius chalybeus chalybeus (Hemprich&Ehrenberg). 1 d^, Fazogli, Blue-Nile, 10 Jan. Common in small flocks around villages. Lamprocolius chloropterus schraderi Neumann. 1 9 , Fazogli, Blue-Nile, 10 Jan. An immature bird; w. Ill, cul. 18, tail feathers, 69, tarsus, 24. Spreo pulcher (P. L. S. Miiller). 1 &, Singa, Blue-Nile, 25 Feb. Only seen a few times; apparently does not occur far south. Oriolus chryseos (Heuglin). 1 9 , Fazogli, Blue-Nile, 19 Jan. PHILLIPS: BIRDS FROM THE SUDAN. 27 The bill of this specimen is shorter and stouter culmen 20 mm. than that of a specimen of 0. auraius from Gambia, in the M. C. Z. Only seen at Fazogli. DiCRURIDAE. DiCRURUS AFER (Lichtenstein). 1 cf and 3 9 's, Sennar, Blue-Nile, 25 Dec. ; Singa, 27 Dec. ; Ga- bardi, 31 Dec; Magangani, 29 Jan. Very common. CORVIDAE. CoRvus SCAPULATUS Daudin. 1 9 , El Mesharat, Blue-Nile, 4 Jan. Very common. CORVUS CORXIX SHARPEI Oatcs. 1 9 , Cairo, Egypt, 12 Dec. Seen only as far up the Nile as Luxor. Rhinocorax affinis (Riippell). 1 cf , Fazogli, Blue-Nile, 18 Jan. Only seen around Fazogli. The following Publications of the Museum of Comparative Zoology are in preparation : — LOUIS CABOT. Immature State of the Odonata. Part IV. E. L. MARK. Studies on Lepidosteus, continued. " On Arachnactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II. H. L. CLARK. The "Albatross" Hawaiian Echini. with 14 Plates Reports on the Results of Dredging Operations in 1877, 1878, 1879. and 1880, in charge of Alexander Ag.\.ssiz, by the U. S. Coast Survey Steamer "Blake," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. The Crustacea of the "Blake." A. E. VERRILL. The Alcyonaria of the "Blake." Reports on the Results of the Expedition of 1S91 of the U. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tanner, U. S. N., Commanding, in charge of Alex.ander Agassiz, as follows: — K. BRANDT. The Sagittae. The Thalassicolae. O CARLGREN. The Actinarians. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. " The Schizopods. BAROLD HEATH. Solenogaster. W. A. HERDMAN. The Ascidians. S. J. HICKSON. The Antipathids. E. L. MARK. Branchiocerianthus. JOHN MURRAY. The Bottom Speci- mens, of P. SCHIEMENZ. The Pteropods and Heteropods. THEO. STUDER. The Alcyonarians. The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. R. V. CHAMBERLIN. The Annelids. Reports on the ScientiQc Results of the Expedition to the Tropical Paciflc, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Com- manding, as follows: — H. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. •S. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. H. LUDWIG. The Starfishes and Ophi- urans. ■G. W. MtJLLER. The Ostracods. MARY J. RATHBUN. The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes. T. W. VAUGHAN. The Corals, Recent and Fossil. R. V. CHAMBERLIN. The AnneUds. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT. HARVARD COLLEGE. There have been published of the Bulletin Vols. L to LIV.; of the Memoirs, Vols. I. to XXIV., and also Vols. XXVI. to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, and XLI. Vols. LV. to LVIII. of the Bulletin, and Vols. XXV., XXX., XXXV., XXXIX., XL., XLII. to XLVIIL of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation : — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S. N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905^ Lieut. Com- mander L. M. Garrett, U. S. N., Commanding. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. Contributions from the Geological Laboratory, Professor R. A. Daly, in charge. These publications are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology, Cambridge, Mass. 3\**^^ Bulletin of the Museum of Co:aai>^xo Mem. M. C. Z., Vol. XXXV, No. 2, August, 1907, 56 pp., 9 pis. " Bull. M. C. Z., Vol. LI.. No. 6. November, 1907, 22 pp., 1 pi. 12 Bull. M. C. Z., Vol. LIL, No. 1, Jime. 1908, 14 pp., 1 pi. 13 Bull. M. C. Z., Vol. LIL, No. 2, July. 1908, 8 pp., 5 pis. 14 Bull. M. C. Z., Vol. XLIIL. No. 6. October, 1908, 285 pp., 22 pis. t» Bull. M. C. Z., Vol. LIL, No. 5, October, 1908, 11 pp.. 2 pis. " Mem. M. C. Z., Vol. XXXVII., February. 1909, 243 pp.. 48 pis. " Mem. M. C. Z.. Vol. XXXVIII., No. 1, June, 1909, 172 pp., 5 pis.. 3 maps. " Bull. M. C. Z.. Vol. LIL, No. 9, June, 1909. 26 pp., 8 pis. »» Bull. M. C. Z., Vol. LIL. No. 11, August, 1909, 10 pp., 3 pis. 20 Bull. M. C. Z.. Vol. LIL, No. 13, September. 1909, 48 pp., 4 pis. 21 Mem. M. C. Z., Vol. XLL, August, September, 1910, 323 pp., 56 pis. 52 Bull. M. C. Z., Vol. LIV., No. 7, August, 1911, 38 pp. » Mem. M. C. Z., Vol. XXXVIII., No. 2. December. 1911. 232 pp., 32 pis. 24 Bull. M. C. Z., Vol. LIV.. No. 10, February, 1912, 16 pp., 2 pis. 2» Mem. M. C. Z.. Vol. XXXV., No. 3, April, 1912, 98 pp., 8 pis. « Bull. M. C. Z., Vol. LIV., No. 12, April, 1912, 38 pp., 2 pis. *' Mem, M.C. Z.. Vol. XXXV. No. 4, July. 1912, 124 pp.. 12 pis. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE. Vol. LVIII. No. 2. EXPLORATIONS IN THE GULF OF MAINE, JULY AND AUGUST, 1912, BY THE U. S. FISHERIES SCHOONER GRAMPUS. OCEANOGRAPHY AND NOTES ON THE PLANKTON. Bt Henry B. Bigelow. With Nine Plates. [Published by Permission of H. M. Smith, U. S. Fish Commissioner.] CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. February, 1914. No. 2. — Explorations in the Gulf of Maine, July and August, 1912, by the V. S. Fisheries Schooner Grampus. Oceanography and Notes on the Plankton. By Henry B. Bigelow. CONTENTS. Bav The cruise .... Equipment and methods Oceanography Surface temperature Temperature sections . Temperature at 25 fathoms Bottom temperatures . Temperature profiles Seasonal changes in Massachusetts Salinity ..... Surface salinity Salinity at intermediate depths Salinity at 25 fathoms Salinity at 50 fathoms Bottom salinity Salinity profiles Density Color .... Transparency Circulation in the Gulf of Maine Tidal currents Surface and bottom currents Circulation as shown by temperature and salinity Comparison with previous records of temperature and General considerations ..... Preliminary notes on the macroplankton . List of fishes ....... Larval and postlarval stages Adult stages ...... List of copepods ...... Distribution of the more important plankton species Calanus finmarchicus ..... Euchaeta norvegica ..... Anamalocera pattersoni .... salinity 32 35 42 43 48 54 55 56 58 62 62 64 6S 70 70 71 76 81 82 83 83 85 87 92 95 98 107 107 111 115 117 117 118 118 32 bulletin: museum of compaeative zoology. Meganyctiphanes norvegica Euthemisto compressa . Clione limacina Limacina balea Salpae .... Tomopteris helgolandica Chaetognaths Medusae Siphonophores Ctenophores Hesults of the quantitative hawls Microplankton Table of stations Table of temperatures Table of salinities Table of densities Table of current measurements Bibliography Explanation of plates. PAGE. 118 119 119 120 121 121 121 123 125 126 127 130 135 137 137 141 143 145 The cruise. During July and August 1912 the U. S. Fisheries Schooner Gram- pus was detailed for an oceanographic cruise in the Gulf of Maine, under my direction, the purpose being to make as nearly complete a survey of the temperatures, salinities, currents, and plankton, of the waters of the Gulf as the brief time at our disposal, anfl the limitations incident to the use of a sailing vessel would allow, (Bigelow, 1913). It was also planned to do some systematic trawling in the neighbor- hood of Casco Bay, in cooperation with the Harpswell Marine Lab- oratory. During the cruise I was accompanied by Messrs. W. W. Welsh and Herbert E. Metcalf as assistants. It is a pleasure to ac- knowledge the assistance which Dr. C. O. Esterly has afforded in the preparation of this report, by identifying the copepods in more than 60 samples of plankton, no small task. And the value of the dis- cussion of the plankton (p. 98) is largely due to his efforts, for copepods were altogether its most important constituent. A like debt of thanks is due to Mr. E. L. Michael, who has identified many of the Sagittae (p. 121), and to Mr. W. W. Welsh, who supplied the lists of fish fry and adult fishes (p. 107). I am also indebted to Capt. John W. McFar- land, of Gloucester, who made several "tows" from his Schooner Victor. RIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 33 Up to the present time, \eiy little attention has been paid to the oceanography of the Gulf of Maine. But the fact that waters of diametrically opposed origins, *'. e. Gulf Stream water and cold coast water, have long been known to meet each other here, gives reason to expect that an examination by modern methods will be of general oceanographic interest, and ma}' be expected to have a practical bearing on the extensive fisheries of which it is the seat. It is obvious that observations restricted to two months in mid- summer can not afford a picture of the regular series of changes which its waters undergo during the year, or of the sporadic variations which may be expected from the geographic position of the region in question, and from its relation to the Gulf Stream. Consequently the following report is to be regarded only as the beginning of a survey which, it is hoped, will be continued at other seasons in ensuing years. The preparations for the cruise were made in Gloucester, and our first station, some five miles off that harbor, was occupied on July Qdi, when we made a trial of the winches, trawl, deep-sea thermometers, water-bottles, and of the current-meter. The current measurements must, of course, be made from a boat at anchor; and we found that time was economized by taking them, and the serial temperatures and w'ater samples as well, from a dory, which we could easily anchor in any depth of water down to 150 fathoms. Our first field of w'ork was the northern part of Massachusetts Bay. We then ran out to the 100 fathom basin, some 35 miles east of Cape Ann, where we made Station 7; but unfortunately the sea was so rough that it was impossible to make a quantitative haul, although the other work, including the hydrographic observations from the dory, was successfully performed. The nature of the hauls and other observa- tions made at this and the other stations is tabulated below (p. 135). From the 100 fathom basin we ran in toward Ipswich Bay, where the plankton is proverbially rich, making a rich trawl-haul of fishes at Station 8, and taking observations in the deep trough between Jeffrey's Ledge and the coast. At Station 10, off Portsmouth, our trawl fouled in some obstruction, and the winch failed to pay out the wire rope, with the result that we lost the trawl with 150 fathoms of wire rope, broke the dredging boom, and did so much damage that we were forced to return to Gloucester to refit. After the damage was repaired, heavy weather delayed us until July 22d, when we ran northerly to Casco Bay, touching at Ports- mouth, and occupying Stations 12-14, to develop the hydrographic conditions along the coast and in the trough west of Jeffrey's Ledge. 34 bulletin: museum of comparative zoology. According to previous agreement Casco Bay was made our headquart- ers until July 31st (Stations 15-20), the vessel being engaged in dredg- ing and trawling in the Bay and oflF its mouth, in cooperation with the South Harpswell Marine Laboratory. On the completion of this work, July 31st, the vessel proceeded along the coast as far as the mouth of Penobscot Bay, making one offshore Station (21), and numerous hauls in the coastal waters and among the islands, while I remained at the South Harpswell Labora- tory and titrated all the water samples collected up to that date, a room being placed at my disposal by the Director, Dr. J. S. Kingsley. I rejoined the Grampus at Portland; but owing to heavy weather and thick fog, it was not until August 7th that we were able to resume work. We now ran a triangle to Piatt's Bank and Jeffrey's Bank, likewise making a station off Cape Elizabeth, one in the deep trough between Piatt's and Cash's Ledge, and one between Jeffrey's and the mouth of Penobscot Bay; but on the evening of August 8th, we were driven to refuge in Boothbay by thick fog, and lay storm-bound there and in Portland Harbor for a week. Leaving the latter port on August 13th, we commenced a section toward Cape Sable, following the paral- lel of 48° 25', making Stations 27 and 28 in the eastern part of the 100 fathom basin, and Stations 29 and 30 on German Bank off the Nova Scotia Coast on the evening of August 14th in thick fog. The follow- ing day Station 31 was occupied off Lurcher Shoal, the exact position doubtful because of the fog. That afternoon we spoke a fishing vessel lying at anchor on the Grand Manan Bank and making a good fare of cod ; during the night the fog lifted, allowing us to pick up the light house on Petit Manan Island. At daylight, August 16th, the weather having cleared, we occupied Station 32, some ten miles off Mt. Desert Rock, and then turned northeasterly along the coast, making a station off Moose Peak. That night we made Station 34 in the Grand Manan Channel, and anchored in Eastport the following morning. On our passage through the channel we had found almost no plankton, a result in very marked contrast to the hauls which we had made off shore and further to the west (p. 104) ; and our run homeward was planned to develop the limits of this barren area as well as to trace the breadth of the band of cold water which lies close to the coast of Maine. Consequently on leav- ing the Grand Manan Channel, August 20, we ran off shore once more to the 100 fathom basin (Station 36) where we found an abundant plankton, and then turned northward again, reaching the coast near BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 35 Mt. Desert, whence we followed the outer islands (Stations 37-39) to the mouth of Penobscot Bay. On August 21st heavy fog once more set in, and on the 22nd we were driven to refuge until the 24th, in the Kennebec River, whence we ran direct for Cape Ann. We had planned several stations for this nm, but heavy sea so interfered with our work, that only surface and intermediate hauls, bottom temper- ature, and water sample were taken at one station. Up to this time we had been covering fresh ground constantly, thus having little chance to trace the changes in hydrographic con- ditions consequent on the advance of the season. But we were now able to repeat in ^Massachusetts Bay some of the stations which we had occupied sLx weeks earlier. One Station (43) was likewise occupied off Cape Cod, and on August 31st the Grampus returned to Gloucester. Equipment and methods. The mone}^ available for fitting the Gr.aivipus for the cruise was limited, and we were therefore unable to provide ourselves with various pieces of apparatus which would have been desirable. The Grampus has no dredging engine, to remedy which deficiency a gasoline winch, built for her on a previous occasion (Bigelow, 1909), was installed on deck just forward of the mainmast. But as this machine has a cargo- drum only, it was necessary to wind the wire rope from it by hand on a second winch. The reeling drum carried 300 fathoms of plough-steel rope, I in. in diameter, with which all the trawling, dredging, and tow- ing with the large horizontal and vertical nets was done, the length of wire outboard being measured by a fathom recording sheave. A small hand winch with divided barrel carrying 300 fathoms of soft iron rope ^ in. in diameter with breaking strain of 500 lbs., and 400 fathoms of malleable steel sounding wire was also used. The little winch was used in the dory, for serial temperatures, serial water samples, and current measurements; and occasionally on the vessel for similar purposes. "Soundings were usually made by hand with cod-line and 30 lb. lead, a method sufficiently accurate for depths of less than 150 fathoms; but occasionally with the | in. wire, or with the sounding wire. The surface thermometers were of two kinds; the ordinary " Bureau of Fisheries" type (Tanner, 1897) graduated to 1° F, and a set of six extremely accurate chemical thermometers provided by R. Goertze, Leipzig, graduated to .1°C. Most of the observations were made 36 bulletin: museum of comparative zoology. with the former, as the readings are sufficiently accurate for the pur- pose, and they are much more convenient in actual use. Two of them were used, their rating being so close that there was no appreci- able difference between them. We carried four Negretti and Zambra reversing deep-sea thermom- eters, unfortunately without auxiliarj^ thermometers for taking the temperatures of the detached thread at the moment of reading, such as are provided in their latest pattern and in the Richter thermometer. Two of these were rated in the U. S. Bureau of Standards at Washing- ton, two in the Chemical Laboratory at Harvard University, with the following results: — Negretti and Zambra Thermometer, U. S. B. F., IVo. 7,277. Reading, °F Correction, °F when T. of detached thread is 32° 60° 90° 32° -.3° -.5° - .6° 60° -.6° -.9° -1.1° 90° 0 +.3° -f .6° Negretti and Zambra Therviometer, U. S. B. F., No. 7,259. 32° -.5° 60° -.2° 90° +.4 o -.8° -1.1° -.5° - .9° 0 - .5° It is fortunate that the changes in reading consequent on change of temperature of the detached thread are so small, for without the use of a M'ater-bath, which was not available, the temperature of the detached thread could be obtained only by allowing the instrument to come to the temperature of the air before reading. The corrections for Nos. 84,036 and 49,648 were noted with the temperature of the detached thread the same as that of the readings ; i. e., the freezing point reading was taken at an air temperature of 32°, the 68.5° reading at an air temperature of 68.5°. They are as follows : — 84,036, 32°, correction -.2°; at 68.5°, -.55°; at 77.13°, -.82°. 49,648, 32°, correction -.5°; at 68.5°, -.16°; at 77.13°, -.37°. With both these thermometers the requisite correction at readings between 40° and 50° is about — .3°: and though this is not exact, varia- tions from it, within this range, are less than the probable error of BIGELOW: EXPLORATIONS IX THE GULF OF MAINE. 37 the observations (p. 40). The thermometers were used in reversing cases of the Tanner tj^pe (Tanner, 1897, pi. 21) actuated by a propeller; and these worked very well. Two water-bottles were taken for collecting samples, a "Sigsbee" (Tanner, 1897, pi. 24): and a stop-cock bottle; but as the first trial of the "Sigsbee" showed that it could not be relied upon, all subsequent samples were obtained with the stop-cock bottle. This apparatus is a modification of the stop-cock bottle used on the ]\Iicil\el Sars and highly recommended by Hel- land-Hansen and Xansen, (1909) the chief difference being that it is single instead of double, and actu- ated bv a messenger instead of bv a propeller. In its essentials (fig. 1) it consists of a brass tube, tinned on the inside, with a stop-cock at either end, the openings of the latter being only slightly smaller than the inside diameter of the tube. The mouth of the lower one carries a large copper funnel, which hastens the flow through the tube as it is being lowered and prevents water being carried downward in the bot- tle. Each st<5p-cock is hinged by a rod to the brass plate which carries the tripping gear, in such a way that when the bottle is raised both stop-cocks are open. When the bottle is tripped, the tube falls of its own weight, the hinge-rods turn- ing the cocks in their barrels, and closing them. The tripping gear consists of a scear which engages the end of the upper hinge-rod when the tube is raised, and of a trigger which trips the scear when pushed downward against its spring by the messenger which is sent down along the wire rope. The dog, or ratchet engages the lower hinge-rod when the bottle falls and is closed, to prevent accidental opening. Tliere^is^a Fig. 1. — Stop-cock water bottle. 38 bulletin: museum of comparative zoology. small stop-cock near the upper end to admit air, and another near the •lower end to discharge the water. The apparatus proved entirely reliable, perfectly water tight, and it has the great advantage that it can be made by any skilled machinist at small expense. The most important precautions in its manufacture are to pro\ide tight stop-cocks : and to make the diameter of the tube as nearly the same as that of the latter as possible. The water samples were preserved in "citrate of magnesia" bottles, made of lead glass by the Whitall Tatum Co., with patent stoppers consisting of a porcelain disc forced by a spring against a rubber ring. The joint thus formed is so nearly air tight, that the danger of evap- oration is negligible. As pointed out (p. 62) tests show no appreciable alteration of the samples after prolonged storage. The only drawback to these bottles is that they are fragile and occasionally break sponta- neously as a result of sudden change of temperature. Current measurements were taken with an Ekman current meter. Salinity Avas determined by titration with nitrate of silver, the index being chromate of potassium. The burette and "Knudsen" 3-way pipette were supplied by Robert Goertze of Leipzig, the standard water by the International Committee for the exploration of the sea. This, of course, is the method almost universally employed ; and the principle on which it depends has been explained by Murray and Hjort, (1912) as well as by various other writers. The color of the sea is usually recorded by the "Forel" scale based on a combination of blue and yellow, the former being .5 gram copper- sulphate + 5 cc. ammonia in 95 ce. w^ater, the latter .5 gram potassium chromate in 100 cc. water. The combinations used are: — 1 2 3 4 5 6 7 8 9 10 11 12 13 blue 100 98 95 91 86 80 73 65 56 46 35 23 10 yellow 0 2 5 9 14 20 27 35 44 54 65 77 90 In practical use a scale consisting of a series of glass tubes- is unsatis- factory because of surface reflections. But these are entirely avoided if the tubes be mounted in a frame above a white mirror of porcelain at 45°, being thus seen by transmitted light against a white back- ground. The color of the sea water is observed by means of an ordi- nary plate-glass mirror mounted at 45° at the end of a pole and held a foot or two below the surface on the shady side of the ship. With this device, our home waters change from apparent blue to light bottle- green. Transparency measurements were made with the ordinary white BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 39 disc fourteen inches in diameter, and we likewise used a four candle- power electric light with storage battery, in a water-tight brass case with glass window at the top. The following nets were used: — 1. Four foot open net for horizontal towing, of the Albatross pattern; ten ft. long, the upper five ft. with i in. mesh, the lower five ft. lined with silk, 38 meshes to an inch. A glass bucket was some- times used with this net, and a 70 lb. weight attached to the wire rope. 2. Quantitative nets of the Hensen type, the opening of the net 36 cm. in diameter, with glass collecting-bucket, and a 70 lb. weight attached to the latter. Nets of two grades were used, the silk of one being 74 meshes to the inch, the other 144 to the inch. 3. Ordinary open net of no. 20 bolting silk, 18 inches in diameter. 4. Open net 12 inches in diameter, silk 38 meshes to the inch. 5. A scrim net 18 inches in diameter. 6. A closing net for horizontal towing. This net, described in Int. rev. hydriob., 1913, 5, p. 576, is a com- bination of the Chun-Petersen-Nansen principles, i. e., it has a hinged ring which is sent down closed, to be opened by a spring released by Fig. 2. — Closing net in operation. a messenger; and it is closed by a draw-string about the net bag, likewise operated by messenger (fig. 2). . i • Trawling and dredging were a minor part of our program; for this work we carried ordinary dredges, and eight-ft. beam trawls. The four-ft. and closing nets were towed horizontally, sometimes separately, sometimes simultaneously on the wire rope. In the latter case, the former was necessarily used at the deepest, the latter at the intermediate horizon. In the shallow waters in which we worked the catenary of the rope is so small as to be practically negligible; 40 bulletin: museum of comparative zoology. and the depth can be calculated from the angle of the rope as observed by the dredging quadrant (Tanner, 1897) and the length outboard. The need for a high degree of accuracy in oceanographic research has been emphasized by Helland-Hansen and Nansen, (1909) who have shown that in waters as comparatively well known as those of the North Sea and the Norwegian Sea, inaccurate salinity observations are worse than none, as they give a wholly misleading idea of the water- circulation. The same is true also of temperature readings, especially at great depths. But in a preliminary^ survey of a field, so little known as the Gulf of Maine, the same high degree of accuracy is not so essen- tial, for any information which can be relied on as approximately correct is of value. Nevertheless, the more accurate the determina- tions the better, for the sake of future comparisons. In any case, it is essential that the probable limits of error of the obser^'ations for both salinity and temperature should be clearly stated, and constantly borne in mind in all discussions. In the determinations for salinity we are pro\'ided with a perfectly satisfactory water-bottle; the storage of the samples is not open to any apparent criticism, and our burette and pipette are of the best. The instrumental error, therefore, must be very small indeed; and there remains only what we call the personal error of the observer. Unfortunately no trained chemist was available for the titrations;^ and I must confess that I have found the determination of the precise point at which the color changes from yellow to orange a difficult one. Nevertheless, as every sample was titrated twice, some of them three or four times, as the standard water could be relied upon, and as an actual test (p. 62) has shown that repeated tests of the same samples did not differ by more than .01 of salinity, I believe that the results arrived at are reliable considerably within the requirements of the International Committee for the exploration of the sea, i. e., ^.05 of salinity, probably to =•= .02 of salinity. In the case of temperature, a very high standard of accuracy could not be expected from the instruments which we used. Our deep-sea thermometers were graduated only to l.°F; and the graduations are so rough that we found it impossible to rely on estimation closer than .2°F, though the readings were taken with a reading lens, and estima- tion to .1°F was constantly attempted. We must also consider the possibility of error resulting from not knowing precisely the tempera- ture of the detached mercury thi'ead when read, though the table of correction shows that an error here of 5°F, at the usual air temperature of 55°-70°F would make a difference of only about . 1°F in the reading. BIGELOW: E^IPLORATIONS IN THE GULF OF MAINE. 41 and this maj' be considered the extreme. There is one other source of error in any reversing thermometer actuated by a propeller; i. e., uncertainty at what precise level the instrument reverses, with possi- bility of change in reading during its passage upward through the column of water necessary to reverse it. But we so often used two thermonieters at each level, and so often repeated the entire series, that I do not believe this possible error is of any practical importance in the present case. On the whole, then, it is better not to claim accu- racy closer than ±.3°F; i.e., roughly, .15°C. And it is certainly much better to set these limits wide, rather than to claim a higher degree of accuracy than can be relied upon. The surface thermometers were extremely reliable, and so far as the instruments themselves are concerned very little error is to be expected. But the readings were taken by various persons, often under difficult conditions, therefore accuracy is not claimed beyond ± .5°F. C>pc Eluahtthj Fig. 3. — Bathymetic chart of the Gulf of Maine. 42 bulletin: museum of comparative zoology. Oceanography. Up to the present time no systematic studies of the oceanography of the Gulf of Maine have been undertaken. The surface tempera- tures have, of course, been known in a general way for many j^ears, as has the existence of a cold band of water close to the coast of Maine and in the Ba^' of Fundy; and thanks to Dickson's, (1901) researches we have a fairly satisfactory idea of the seasonal range of surface tem- perature for two years, 1896 and 1897. But his records were far too few to delimit the distribution of slightly differing temperatures within the Gulf. Almost all the knowledge we possess as to the bottom temperatures dates back to 1872, 1873, and 1874 when a series of dredgings was carried out by the U. S. Fish Commission and the U. S. Coast Survey on George's Bank, in the Bay of Fundy, off Cape Elizabeth, and at various other localities in the Gulf. The bottom temperature was re- corded at each station, and the records have been published by Yerrill, (1873-1875); but unfortunately, as he himself points out, the Miller- Casella thermometers which were used proved unreliable, two instru- ments often differing by several degrees when used simultaneously. Nevertheless the results were valuable as showing in a general way the low bottom temperature of the Gulf (p. 93). So far as I can learn, no intermediate temperatures have ever been taken in the Gulf, except a few which I obtained during the summer of 1911 between Cape Ann and Casco Bay. The salinity records for the Gulf are even more scanty than those for temperature. A considerable number of hydrometer readings for the surface have been taken b v the Bureau of F'isheries ; but most of them were made with unstandardized instruments, and under circum- stances precluding any approach to accuracy. The only reliable salinity records from the surface are three titrations by Dickson, (1901), of samples collected oft' Cape Cod, April, 1896; off Cape Sable, April, 1896; and northeast of George's Bank, April, 1896. And there are no records whatever of the salinity on the bottom, or at inter- mediate depths. For George's Bank and the Eastern Channel, the data is rather more extensive, there being eighteen titrations (Dickson, 1901); and a considerable series of temperatures were taken by the Albatross in 1883 in the channel with Negretti and Zambra reversing thermometers. There is one titration from Brown's Bank and a considerable number BIGELOW: EXPLOKATIONS IN THE GULF OF MAINE. 43 southeast of Nova Scotia (Dickson, 1901) besides a series of surface and bottom temperatures by the Albatross (Townsend, 1901). Surface temperature, JnJij- August, 1912. — The surface temperature was taken hourly, day and ni<,dit, throughout the cruise; and the read- ings are plotted on the chart (Plate 1). When I came to check up the results, one interesting anomaly became apparent, namely, that the surface temperature at each station is from .5° to 1° lower than the next reading on either side of it. This discrepancy is probably due to the method of observation, the rearlings at the stations being taken with the tlierraometer hanging a foot or so below the surface, whereas the instrument dragged on the actual surface when the vessel was under way. The chart shows that so far as surface temperature is concerned the Gulf of Maine can be divided into two general regions, one with tem- peratures of 60° F or over, both day and night, in July and August, the other with temperatures below 60°. In a general way the first includes the whole of the southern and central parts of the Gulf, i. e., Massachusetts Bay, and the off-shore waters south of 43° 21' N. Lat., as far east as 66° 45' W. Long., but it does not reach the Nova Scotia coast. Over all this area the daily average of the surface water was about 61° and the diurnal warming, touched on below, consider- able. But though Massachusetts Bay as a whole belongs to the warm division, lower temperatures were observed along the northeast coast of the Bay, near Eastern Point, off Race Point (Station 44, 58°); off Baker's Island, and notably near Boston Light-ship (July 15, 58°) where two days before a temperature of 63° was observed. And on July 23 a band of water of only 56° was found extending from Gloucester around Cape Ann for some ten miles northeasterly, i. e., covering a region where a few days before temperatures above 60° M^ere found. The temperature was above 60° in Ipswich Bay, north of Cap e Ann. But when we entered the passage between the Isles of Shoals and the mainland, the surface temperature dropped several degrees, the readings here being 55°-57°, and working northeastward, a con- tinuous belt of this cold water was found lying next the coast. From the Isles of Shoals nearly to Cape Elizabeth this cold band was about 15 miles broad; south of the Isles of Shoals it narrowed suddenly, the 60° curve touching the coast somewhere between Station 10 and the mouth of the Piscataciua River. The cold water does not reach Cape Ann except sporadically, an instance, as noted above, being July 24th, when, strong northerly gales for the three preceding 44 bulletin: museum of comparative zoology. days had driven the warm surface water to the south. And even in this case it is probable that the cold water which took its place welled up from below, rather than that it was an extension of the cold zone normally encountered some 15 miles further north. At 43° 27' N. Lat., i. e., a few miles south of Cape Elizabeth, the cold band suddenly became broader, the 60° curve bending eastward almost at a right angle, and roughly following the parallel of 43° 27', to within about 35 miles of Seal Island, Nova Scotia {i. c, 66° 49' W.) where it turned southward and passed out of the area covered by the cruise of 1912. The cold water thus expands from a narrow band to a triangular area which is about 45 miles broad opposite Grand Manan. It is continu- ous thence along the western coast of Nova Scotia, becoming narrower again (25 miles broad) off Yarmouth. Throughout this triangle the temperatures, day and night, were everywhere 59° or below, except for one sporadic reading of 60° off the Grand Manan Bank, probably explicable by diurnal warming on a very calm day; and the diurnal range very small. From Portland eastward to Mt. Desert the tem- perature range was from 56°-58°, a very small variation when we remember the strong tides of this region. Northeastward from Mt. Desert the temperature close to the coast dropped below 55°; and from Moose Peak to and through the Grand Manan Channel, as well as in Passamaciuoddy Bay and Eastport Harbor the temperature on the surface was 50°-52°. Unfortunately we did not enter the Bay of Fundy proper, and it is therefore impossible to draw the curve of 55° accurately. But so far as our observations show, it touched the outer islands at Mt. Desert; ran easterly for about 25 miles, and then turned southeasterly, en- closing the Bay of Fundy and a band along the west coast of Nova Scotia. On our run from Station 28 to Station 29, the drop in tem- perature was very sudden, from 60° at 60° 49' W to 50.5° at Station 29, 20 miles further east. The area of water colder than 55°, is then roughly comparable in outline to that between 55° and 60°, though much smaller in extent. And this cold water was below 55°, usually below 53°, by day as well as by night. The lowest surface tempera- tures encountered were on German Bank (50.5°) off Grand Manan Bank (50°), and in the Grand Manan Channel (50°). Our only example of seasonal change is in Massachusetts Bay, which we studied at the beginning and again at the end of our cruise. From July 9-15 the temperatures in the northern half of the Bay during the day time were usually 63°-65° (60°-65°) except for the occasional cold bands mentioned (p. 43) to which we will have occa- BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 45 sion to return in our discussion of vertical circulation; and off Cape Ann the temperature during this same period ranged from 60°-66° (day and night) ; usually r)3°-65° in the day time. On our return we crossed Massachusetts Bay twice (August 28th-31st). On the first passage the surface temperature ranged from 60°-62°, the mean being about 61°; on the second, two days later, from 59°-61°, the mean being nearly 61°; and on August 29th, off Cape Cod, the temperature range was from 60° to 62°, with a mean of 61°. These observations show that by the end of August an appreciable cooling of the surface water had taken place in and near Massachusetts Bay, from the annual maximum, which must be reached about the first of August. Satisfactory data as to diurnal warming can be obtained only when the vessel lies at one spot for considerable periods, so our information on this point is not very extensive. But we made some observations which suggest an unusually great diurnal warming under certain con- jj ^ A.M. 3 9 10 1 I 12 2 ; 5 A \ P.M. 5 6 7 8 9 10 1 A.M. 12 1 70 f 69 68 k--( — ,-' -. "', 67 66 65* 64 63 62 \ / \ \ } \ \ / . \ ,^ ^ \ / \ 01 60' 59 / \ 1 Fig. 4. — Air and surface temperatures, off Cape Ann, July 15, 1912. ditions. On July loth we ran eastward from Massachusetts Bay to Station 7, and then westward again in the evening, being continu- ously within an area of weak tides, with clear sky and moderate breezes. Surface and air temperatures for each hour from 7 A. M. to 12 midnight are shown (fig. 4). The surface temperature, which was 60°, near Boston Light-ship, rose rapidly to 63° at 10 a. m. It then remained constant until 2 p. m., when there was an irregular rise, culminating, at 7 p. m., with 66°. After this the temperature fell reaching 60° once more at midnight. Obser^•ations made during the rest of the night are not comparable with the foregoing, because we 46 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. were then within a few miles of the coast; but they show that the temperature remained 60°-61° until 8 A. M., then rose gradually to 67° at 1 p. M., July 16, at which time we were in Ipswich Bay. In the afternoon we passed into the cold coast water off Portsmouth. The air temperature for July loth shows a rise and fall roughly parallel to that of the water, the latter, however lagging far behind the former. On the 16th the air temperature rose from 64° at 6 A. M. to 76° at 11a. m., i. e., it was roughly parallel to the rise of the water. On August 7th we had a second opportunity to observe diurnal warming of the surface. This day was flat calm, with a bright sun, but slightly hazy. We ran all day southeastward from Cape Eliza- beth. Close to the coast, of course, we passed through the cold band; but at 9 A. M. we had run into the warm off-shore water, some fifteen miles from the Cape; and air and water temperatures for every hour from this point on until midnight are plotted (fig. 5). The surface i < A.M ) 10 1 i 12 2 . 5 4 I P. M ) 6 ■ r 8 < 3 10 II 12 2 67' 66 65 64- 65 62 61 60° 59 58 / y / Y \ / I-— 1 /'\ 1 — — , — ,'' ^ *\ L 1 '/ X ' \" > ( / / Fig. 5. — Air and surface temperatures, August 7, 1912. temperature rose steadily from 60°, until at 1 p. m. the maximum, 67°, was reached. By this time the air temperature had risen only 1° (from 63°-64°) ; but by 3 p. m., when the water had fallen to 64°, the air reached its maximum for the day, 66°. From this time onward both air and water cooled, until at midnight both were 61°. This case is especially interesting, because the warming of the water pre- ceded that of the air, and reached a higher degree. So far as they go, these observations show that diurnal warming in the region in question is very considerable in clear, calm weather, even as much as 6° or 7°, but it is usually much less, i. e., 2° to 3°. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 47 One day, August 21st, throws light on the diurnal warnn'ng of the cold coast water between Mt. Desert and the mouth of Penobscot Bay. The hourly diagram (fig. (j) shows that there was only about 2° A.M. P.M. 6 r 8 9 10 II 12 2 3 4 5 6 7 8 9 10 II 12 64° 63 62 61 60' 59 58 f 57 56 55* _^ ►.—- . ►- 1 ► < L :z: Fig. G. — Air and surface tempera tui-es, August 21, 1912. ( ) ? ' i. AM 9 l( ) 1 1 12 1 2 -2. 4 r; PM 6 7 ' 8 9 1( ) 1 1^ . 1 2 3 Af 4 v1 5 C ) 7 /■o* 68 67 66 65- /LA ; 1 I 1 * t / 04 63 62 f' '\, / ~-., / 61 60' 59 / ! ': \ hi 57 56 55- \ / \ — —- ^~^ "% \ \ \ / \ ', \ / \ ', 54 53 52 ri \ / \ \ \_ \ / \ \_ 51 50* \ 1 ■w Fig. 7. — Air and surface temperatures, August 6, 1912. 48 bulletin: museum of comparative zoology. rise in the surface temperature readings (55°-57°) although the air rose from 58°-63°, and the temperature readings taken on various days show that diurnal warming is very much less in this region than it is in the warmer off-shore waters. So far as our observations go, they suggest that in the cold coast water northeast of Mt. Desert diurnal warming is not usually observal)le; thus the diagram for August 6th (fig. 7) shows a slight fall (56°-57°) from 6 a. m. until Rth 0 5 10 15 20 25 30 55 40 45 50 55 60 65 70 75 SO 85 90 95 100 105 41° 42 45 44 45° 46 4? 48 49 50" 51 52 55 54 55° 56 S7 58 59 60° 61 62 63 64 65° f ^ -^ . . — — - — — ^ J ^^ ^^ -— - ^ ^"-^ r y y / ^ .it y 2S f M / y y I / y '2S. 24 lz9 / 1 / J / 1 ... Fig. 8. — Temperatui-e sections at Stations 2, 23, 24, 25, 29. noon; although between 9 A. M. and 4 p. m. the air temperature rose from 56° to 68°. To explain the distribution of the surface temperatures of the Gulf of Maine, just outlined, requires a knowledge of the temperatures in the underlying water layers at the same season, which is afforded for the first time by the Cruise of 1912. Temperature sections. — The section made off the mouth of Massa- BIGELOW: EXPLOKATIONS IN THE GULF OF MAINE. 49 Fafk 'J 5 10 15 20 25 30 55 40 45 50 4 1° 42 43 4 4 4 5"46 4 -7 4 8 4 ■9 50° SI 52 53 54 55°56 57 58 59 60" 61 62 65 64 65° — ■ — -^ ^ -^ ^ ^ - ^ i:::^ — - / r ^ ^ ::^ 6 / ,^ ^ ^ / 44, / / // / / / / 45 Fig. 9. — Temperature sections in Massachusetts Bay, Stations 6, 44, 45. ^'■•40*41 42 45 44 45''46 47 4g 49 50°5I '52 55 54 55° 56 57 58 59 Wb,\ 62 63 64 65* 0 , . ^ ^ !==- ' ■-- --— ■ ■:=>-< J .^ ■^ ,-1 .;;»- ''' 10 /• /• ^ ..-■ > 13 / X ^ y J -/ / ^ — ^ •— ^ ^^ 15 20 25 35 40 45 50 55 60 65 •^ ::::? CS^ '4- ^- ^ / y / 14 / yii / Fig. 11. — Temperature sections in the mouth of Massachusetts Bay (Station 2), and west of Jeffrey's Ledge (Stations 11, 14). being the same as it was at Station 2. In the Bay at the end of August conditions are different, as pointed out below. At Station 1 the temperature curve is practically the same as at Stations 5 and 6, the temperature at the bottom in thirty-five fathoms being 40.6°, very nearly what it is at Station 2 at a corresponding depth. The section in the 100 fathom basin off Cape Ann, Station 7, (fig. 10) shows that the surface layer of warm water was slightly thicker here, the drop from the surface to 10 fathoms being only from 64° to about 53°; and the rate of decrease diminishing slowly until the minimum of 40.3° is reached at fifty fathoms, instead of at thirty- 1 To agree with the station numbers of the U. S. Bureau of Fisheries 10000 should be added to the numbers given in this report, e. g. 10002. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 51 five as at Station 2. Below this level there was no further change of temperature to 125 fathoms. At Stations 9, 11, and 14, (fig. 11) west of Jeffrey's Ledge, the curves agree very well with those for Massachusetts Bay, except that the siu-face temperature of the last two is se\'eral degrees lower, and that at one Station (12b) in the trench, a lower bottom tempera- ture, 39.2°, was recorded. But as this was the only instance of a reading below 40.3°, it is possible that the thermometer recorded in- correctly. Off Cape Cod at Station 43, late in August, the bottom temperature was higher, in this case, 41.3° instead of 40.3°; and as at Staton 2, the uniform bottom water was met at 50 fathoms (fig. 10). In all the western part of the Gulf, there was a bottom la\'er, of vary- ing thickness, and reaching to wathin varying distances of the surface of the water, the temperature of which w-as practically uniform, 40.3°. In the western 100 fathom basin, it was seventy fathoms or more in thickness, and it filled the deep circumscribed basin at the mouth of Massachusetts Bay, as well as the bottom of the deep trough west of Jeffrey's Ledge. But the differences in the temperature in Massa- chusetts Bay in early July and late August (p. 58) show that it is only below fifty fathoms or so that the bottom temperature may be expected to remain fairly constant throughout the year. Above that level, the whole water mass is subject to summer warming and winter cooling. If we compare the temperature sections at successive stations from Cape Ann tow^ard Nova Scotia (Stations 2, 7, 23, 24, 27, 28, 29, figs. 8, 10) we find that the curves, wdiich are nearly uniform from the Cape to Station 24, grow progressively straighter from that point eastward, the temperatures being higher and higher on the bottom, lower and lower on the surface. And while the cur\'es for Stations 27 and 28 show that the lower seventy to eighty fathoms of the eastern arm of the 100 fathom basin, like that of the western one, was filled with a layer of water which shows very little decrease in temperature downward lielow thirty-five fathoms, the bottom water differed from that of the western basin in being decidedly warmer than in the latter, a difference which can not be laid to advance of the season, because on our return (Station 41) we once more encountered bottom water of 40.3°, west of Jeffrey's Ledge; and in being less uniform, for it was slightly warmer at all depths at Station 28 than at Station 27. And while at Station 28 the temperature of the whole mass below thirty fathoms was 45.3°, at Station 27 there was a slow, but constant decrease all the way to the bottom, where the tempera- ture in 100 fathoms was about 43°. On reaching German Bank, we 52 bulletin: museum of compaeative zoology. found that the surface temperature had dropped from 59° to 50.5°; the bottom water on the contrary, had risen to 49.2°, the entire drop taking place within ten fathoms of the surface. Temperature sections from Cape Ann to\A'ard the Bay of Fundy, (Stations 11, 19, 39, and 35, fig. 12, and Stations 8, 14, 15, 21, fig. 13), exhibit a gradation similar to that seen on the line Cape Ann-Nova Scotia, the curves growing progressively straighter and straighter Fail 0 5 10 IS 20 25 30 35 45 50 55 60 65 70 75 80 85 90 . 41° 4 2 A ■ZA 4 45° 46 47 A 8 4 9 5 0* 51 52 53 54 55° 56 57 58 59 60-61 62 65 64 65° / 75' ^ ^ — V c- — <^ '-^ ^ /. "^ 1^ / y ^ y /" / ^ / / / / / y / / ■' / I f / / / 35 19 f I 1 / / f / / / 59 Fig. 12. — Temperature sections. Cape Ann to the Bay of Fundy, Stations 11, 19, 39, 35. toward the northeast. Station 11 is practically identical with Sta- tions 2, 23, and 24; Stations 33, and 35 with Station 29; Station 39 is intermediate. It is interesting to compare the temperature conditions over the three off-shore banks which we ^'isited, Piatt's, Jeffre^^'s, and German, (fig. 14) with one another and with those of the deep b^^sins. The first is about fifty miles northeast of Cape Ann. The surface tempera- ture here was 64°, the bottom, reading in 45 fathoms, 40.8°, and its temperature cur^•e (fig. 14) is almost precisely identical with that of Stations 2 and 11. This, of course, shows that the bank had no dis- turbing effect on the water above it. On Jeffrey's Bank, some thirty- BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 53 five miles south of tlie mouth of Penobscot Bay, the surface tempera- ture was cHstinctly colder than we found it on Piatt's the day before, i. e., 57°; but the bottom, in (iO fathoms, was much warmer, 47.3° RiK. 4|'42 45 44 45"46 47 48 49 50° 51 52 55 54 55^ 56 S7 58 59 60° 61 61 63 64 65° Fig. 13. — Temperature sections ia Ipswich Bay (Station 8), oflf Cape Porpoise (Station 14); off the mouth of Casco Bay (Station 15), and off Monhegan (Station 21). F.ih 41° A2 45 44 45° 46 AT 48 49 50° 51 52 55 54 55° 56 57 58 59 60° 61 62 63 64 5 10 15 20 25 50 55 40 45 50 55 Fig. 14. — Temperature section.s on Piatt's Bank (Station 2.3), .Jeffrey's Bank (Station 25) and German Bank (Station 29). / _^_ ^ ^ _ ■ — " / -^ -^ -y y ^^ / ^ y / / / 7^ / / r — / 29 J il i I instead of 40.8°. The curve for Station 25 shows a rapid decline from the surface down to ten fathoms, in which distance there was a drop of nearly 8°, while from that point downward the decline was slow 54 bulletin: museum of comparative zooioGY. and irregular. On German Bank, some fifteen miles oflF Seal Island, Nova Scotia, the surface reading was 50.5°, the bottom 49.3°, the entire drop taking place in the upper ten fathoms, below which point the temperature was uniform to the bottom. These three banks, then, taken in a series, illustrate precisely the same kind of tempera- ture relation as was exliibited by the coast waters passing eastward and northeastward from Cape Ann, but to a more pronounced degree. The serial temperatures at Station 31 (Fig. 15) are especially in- FaMO'A a 43 44 45"46 47 48 49 50° Si 52 53 54 55° 56 57 58 59 60° 61 62 63 u 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 / / /- / / / / / } / / / /f ^ / J / / 35 / / / ^3. sy / / / / / 1 y J n / ' Fig. 15. — Temperature sections in the Grand Manan Channel (Station 35) ; in the northeast end of the Deep basin (Station 36), and near Lurcher Shoal (Station 31). teresting because of nonconformity with those at neighboring sta- tions, i. c. they are warmer at all depths above fifty fathoms, a phe- nomenon best discussed in connection with the temperature profiles. Temperature at twenty-five fatJioms. — The curves for temperature at twenty-five fathoms (Plate 1) reconstructed from the temperature BIGELOW: KXPLORATIONS IN THK GULF OF MAINE. 55 sections, show that the rehitive distribution of temperatures at this depth was in a general way the opposite of what it was on the surface, the lowest temperatures being encountered in the west, in Massachu- setts Bay, off Cape Ann, and in the trough west of Jeffrey's Ledge, the highest in the east, off the coast of Nova Scotia (Stations 29 and 31 ), and in the Grand Manan Channel. The extreme range, at this depth, was from about 42° to about 51°, the former characterizing the cold area delimited above, the latter encountered only at Station 31. In general the twentj'-five fathom temperature in the northeastern part of the Gulf was between 48° and 49°. The curve for 48° runs southerly from the mouth of Penobscot Bay far enough to include Jeffrey's Bank, then turns northward again toward the coast, which it parallels at a dis- tance of about twenty miles, until the meridian of 67° 25' is reached, when it once more bends to the southward. Off the mouth of Casco Bay there was an isolated area where the twenty -five fathom temper- ature was 48° or higher. And off Cape Cod, Station 43, the temper- ature at twenty-five fathoms was likewise above 48°. The chart for this level is constructed only for July and early August, and our ob- servations show that at least in the western part of the Gulf there is a decided rise in temperature at twenty-five fathoms fi'om July 9 to August 31, the water shown on the chart as 42° warming to 45°^6°. Bottom temperatures. — The curves for the temperatures at the bot- tom (Plate 1) show that, like those for twenty -five fathoms, there was a regular rise, both at corresponding depths and al)solutely, passing northeastward from Cape Ann to Nova Scotia. Thus at Stations 2, 3, 5, 6, 7, 12b, and 24, the bottom temperatures in depths of from 40 to 120 fathoms were constantly below 42°, the minimum being 39.2 at Station 12b in the trench between Jeffrey's Ledge and the mainland. That is to say, in July, the bottom temperature over the western arm of the 100 fathom basin, in the deeper parts of Massachusetts Bay, below say forty-five fathoms in the trench west of Jeffrey's Ledge, and over Piatt's Bank was extremely uniform, usually 40.3°. But as we ran eastward we found higher and higher bottom temperatures, irrespective of depth. Thus, at Station 27, on the western edge of the eastern arm of the 100 fathom basin, in 100 fathoms, the bottom read- ing was 43°; at Station 28, thirty-five miles further northeast, in 120 fathoms, it was 45.5°; at Station 29, on German Bank, in thirty-five fathoms, between 48° and 49°, i. e., only about 1.5° below the surface reading; and some 7° or 8° warmer than the bottom temperature at a corresponding depth in Massachusetts Bay, 6° warmer than at thirty- five fathoms over Piatt's Bank. Successive stations passing north- 56 bulletin: jiuseum of comparative zoology. easterly along the coast from Cape Ann toward the Bay of Fundy show a similar rise of bottom temperature, irrespective of depth. Thus at Station 11, in sixty fathoms, abreast of Portsmouth, the bottom read- ing was about 40.3°; at Station 19, abreast of Cape Elizabeth, in fifty fathoms, 42.3°; Station 39, off the mouth of Penobscot Bay, eighty fathoms, 46°; Station 35, in the mouth of the Grand Manan Channel, forty-five fathoms,. 49.3°; which was only about 1° lower than the surface temperature. Temper apurc profiles.- — The first profile (fig. 16) constructed from the temperature curves, shows the distribution of temperature for July and early August from Boston to Station 29, on German Bank, passing through Jeffrey's Bank, (Stations 6, 2, 7, 24, 25, 27, 28, 29). At the western end of the profile, there is a very thin surface layer of warm water with temperatures above 46° overlying the cold bottom water with a temperature of 40.3°-41°, which fills all the east- ern basin below about forty fathoms. Passing eastward the lower limit of the warm layer, which may be established arbitrarily by the isothermobath of 46°, dips from about five fathoms at Station 6 to fifteen fathoms at Station 2; and in the trough west of Jeffrey's Ledge, it lies at about that same depth. From Station 2 to Station 7 it dips to about twenty fathoms, which level it follows to Stations 23 and 24. At Station 25 a very interesting phenomenon is seen, for here the curves for temperatures above 48° rise nearly to the surface, while that of 46° touches the slope of Jeffrey's Bank at about seventy fathoms. East of the bank the re\erse occurs, the curve of 48° rising to about fifteen fathoms at Stations 27 and 28, the curve of 46° to twenty - twenty-five fathoms at these same stations. At Station 29 there is a distortion of the curves parallel to that on Jeffrey's Bank (Station 25) ; the temperature of the entire water-mass being between 49.1° and 50.6°. Over the eastern part of the profile, the bottom water is less imi- form in temperature than it is in the western, the coldest water (about 43°) being met on the eastern face of the slope of Jeffrey's Bank, while the easterly part of the basin below thirty fathoms is filled with water of about 45.3°. A profile from the basin (Station 28) to German Bank (Station 29) passing through Station 31 (fig. 17) reveals the presence of a mass of warm water on the surface at the latter. Over the first part of this line the cur\es for temperatures between 46° and 54° dip sharply, the former descending from about twenty-two fathoms at Station 28 to about sixtv-five fathoms at Station 31. The curve for 50° dips from BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 57 Cs^ ^ S'-— > e) o es o CD CD sO — CS( NO 'tf U-> U3 CD C3 O £v- CO 0% O cii O O o — cs/ rr> 58 bulletin: museum of comparative zoology. twelve to forty-two fathoms; that for 52° from about five to about twenty-five fathoms. The curve for 54°, however, runs parallel with the surface. But the course of the curves eastward from Station .31 shows that the warm water does not extend to the coast bank. On the contrary, the curve of 50° rises sharply until at Station 29 our serial observations locate it at a depth of only about three fathoms. The warm water at Station 31 can be further delimited bj' a profile across the mouth of the Bay of Fundy (Stations 33, 36, 31, 29, fig. 18) which shows that Station 31 is Avarmer at corresponding depths than either of the other three, Station 33 practically reproducing Station 29, ex- cept that the immediate surface was about .5° warmer. Evidently then there was a mass of water lenticular in section, several degrees warmer, at all depths, than the water either east, north, or west of it, at Station 31. Whether this warm area was circumscribed on the south also, or whether it was continuous with the warm off-shore water, possibly even with the Gulf Stream water, which washes the conti- nental slope, is doubtful. The profiles show that the temperature conditions over Jeffrey's Bank, on German Bank, and off the mouth of the Grand ]Manan Channel are closely related to one another, differing correspondingly from the deeper adjacent waters, in being colder at the surface, warmer at the bottom. The three differ from each other, it is true, in degree, but not in kind. But a profile running southeasterly from Mt. Desert for about fifty miles to Station 28 (Stations 37, 32, 28, fig. 19) shows that there is no spreading of the curves on the slope here, which is probably due to the fact that Station 37 lay in the shallow, partially enclosed waters of Frenchman's Bay, where local seasonal warming no doubt played a greater part than it does further off shore. But a profile running off shore from the mouth of Casco Bay (Station 15) to Station 24 shows a spreading of the curves at the shore end (fig. 20) and a profile from Swan Island (Station 38) to the deep basin near Piatt's Bank roughly parallel in direction to the above, shows much the same temperature conditions, with the difference that at the northerly end, which lies just east of the main entrance to Penobscot Bay, the spreading of the curves is more extreme than it is further west. Seasonal changes in Massachusetts Bay. — Our work over the central and northeastern parts of the Gulf did not last long enough to show anything about seasonal changes, further than that the bottom temper- ature at Station 41 (40.3°) compared with what we found oft' Cape Ann and in the trench west of Jeft'rey's Ledge at the beginning of the trip. BIGELOW: EXPLORATIONS IX THE GULF OF MAINE. 59 Fig. 17. — Temperatxire profile from the eastern basin (Station 28) to German Bank (Station 29) passing tlirough Station 31. Fath. 0 55 36 51.1" 52.5° Fig. 18. — Temperature profile across tlie mouth of the Bay of Fimdy to German Bank. 60 bulletin: museum of comparative zoology. Fig. 19. — Temperatiire profile running southeasterly from Mt. Desert to *^j Station 28. r '5 19 ra. 58" 57° 0 24 64° 54.5' 45>^ 60" c;4' • — ^47.5" 50- 47^ '*""~~'i-^ — 45.2.0 46. 4b. S. 42.8' _^^^^^^ ■ v^S,4.2 .2 V^'^^^^^^^^i^C ^^ "m^ 'At "- »w ^--" "^^^^^ W/^^^^^^ 'w//m. 10 zo 30 40 50 60 70 80 90 100 1(0 Fig. 20. — Temperature profile from the mouth of Casco Bay (Station 15) to Station 24. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 61 revealed no appreciable change in temperature of the bottom water in that region from the middle of July to the 25th of August. But comparison between the serial temperatures in Massachusetts Bay- July 9-13 (Stations 1, 5, 6) and those on August 31 (Stations 44, 45, 46) shows a marked warming of the bottom water down to forty fathoms, though, as pointed out above (p. 44), the surface water had cooled appreciably during the interval between our two visits. Stations 6 and 45 are especially instructive because made within a few miles of each other. The surface temperatures (fig. 9) were 61° at both; but whereas on July 14 the temperature was 43° at ten fathoms, and 41.3° at twenty-seven fathoms, on August 31 the ten fathom tem- perature had risen about 10°, i. e., to nearly 53°. At thirty fathoms there was also a rise; but of only 3°, i. e., to 44.7° the bottom tempera- ture, in forty fathoms, being 43.1°. And the curves for Station 45, if continued downward, suggest that 40.3° would not have been met until a depth of about sixty-five fathoms was reached instead of at forty to forty-five fathoms as in early July. But as we were unable to make stations in the deep parts of the Bay on our second visit, it is impossible to state how far such a reconstruction would be correct, though we can safely say that the whole water-mass over the shallower parts of the Bay down to at least forty fathoms was several degrees warmer at the end of August, than it had been the beginning of July, except for the surface, which was slightly colder. One Station, (43), some twelve miles off Cape Cod, over the inner edge of the deep basin, in ninety-five fathoms, remains to complete our survey of the temperatures. With a surface reading of 60°, the intermediate temperatures at Station 43 below five fathoms were from l°-3° warmer at all depths than they were in ]\Iassachusetts Bay two days later (Stations 45 and 46). The temperature curve (fig. 10) is a regular one, without sudden angles. Comparison with the curve at Station 7 (fig. 10) shows that the bottom water at Station 43 was 1° warmer, 41.3° instead of 40.3°; and that it was not encountered until a depth of eighty fathoms was reached, instead of at fifty fathoms, i. e., it was only fifteen instead of seventy-five fathoms thick. Station 43 was colder at all depths above seven fathoms, warmer at all depths below that level. In considering the diflferences between Station 43 on the one hand, and Stations 2 and 7 on the other, the advance of the season ami con- sequent cooling of the surface must be borne in mind. And this no doubt accounts for the lower temperature down to seven fathoms at the former. But the fact that Station 43 was warmer at all depths 62 bulletin: museum of comparative zoology. than Stations 45 and 46, made almost simultaneously, shows that the discrepancy below seven fathoms between it, and Stations 2 and 7, can not be wholly the result of seasonal change, in the sense of solar warming. Hence it seems safe to say that at Station 43 we encoun- tered a water mass distinctly warmer than the waters west, north, or northeast of it. But of course it is impossible to know whether this warm water would have been encountered off Cape Cod earlier in the season, or whether it had moved thither between the times of our two visits to Massachusetts Bay. Salinity. As pointed out above, titration is, on the whole, the most satis- factory method for determining salinity, (the term salinity meaning the number of grams of solids per kilogram of water) ; and the follow- ing account of the salinities of the Gulf of Maine is based entirely on the values arrived at by this method. Every water sample was titrated twice, some of them three or four times, and to test the possibility that some evaporation or other alteration in the salinity of the samples might have taken place be- tween collection and titration, the titrations for four samples, chosen at random, were repeated after an interval of two months, with the following results : — Station Trial A Trial B 27, surface 32.66 32.66 43 95 fathoms 33.69 33.70 22 45 32.74 32.75 2 60 32.92 32.91 The pairs of salinities agree so closely that there was e\adently no appreciable change as a result of storage. Surface salinity.— The chart of surface conditions in July and Au- gust, 1912 (Plate 2) shows that the salinity was lowest close to the coast, there being a band five to twenty miles broad reaching from Cape Ann northward nearly to Cape Elizabeth where the salinity was below 31.4, while it was highest along the western edge of the Gulf, over the Nova Scotia Coastal Bank (Station 31), where water of 32.84 was encountered. The curves clearly show two distinct masses of water of low salinity intruding into the comparatively salt waters of the central part of the Gulf. One of these was off Cape Ann, where the curves of 31.4, 31.8 and 32, swing far to the eastward. The BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 63 curve of 31.S divides ^Massachusetts Bay lengthwise, reaches eastward as far as longitude 69° 61' W, thus including Station 7, then curves back abruptly to within fifteen miles of Cape Neddick, whence it runs northeasterly roughly parallel to the coast, as far as the mouth of Penobscot Bay, and the curves for values below 31.8 show a similar swing. In this region the lowest off-shore salinities observed were 31.08, at Station 14 abreast of Cape Porpoise, 31.2 off the mouth of Casco Bay, and 31.2 at Station 16, near Seguin. But even lower salinities were found at the mouths of the large rivers, i. e. 30.6 at Station 21a in Penobscot Bay. The second intrusion of compara- tively fresh water was encountered off the mouth of Penobscot Bay, where the curve of 32.4 swings off shore southward for some twenty- five miles; but though relatively fresh, this mass of water was abso- lutelj'^ less so than the waters off Cape Ann, its salinity lying between 32 and 32.4, instead of below 32. The conditions in Massachusetts Bay are complex. Both in July and in August the surface salinities of its central portion were between 31.8 and 32; but along the north shore from Xahant to Cape Ann, much higher salinities were occasionally noted, i. c, 32.14 six miles southeast of Baker's Island on July 15th, while a few miles away (Station 6) the salinity was 31.9 two days previous. At Station 1, off Eastern Point, the salinity was 32.07, while at Station 2 it was only 31.7. At Station 44, the only one in the southern half of the Bay, it was likewise higher (32.03) than at the stations made on the same day in the central and northern part of the Bay, the salinity at Sta- tion 45 being 31.9, at Station 46 only 31.6. The curves show, further- more, that while the comparatively saline water of the southern half of the Bay may have been directly continuous, on the surface, with the salt off-shore waters, the high salinities noted along the north shore were isolated patches enclosed by fresher water, i. e., by the curve of 31.8. This phenomenon is important in connection with the fact that it was at just these same localities that abnormally low temperatures were recorded (p. 43). Its significance will be dis- cussed later (p. 90). The salinity of the surface w^aters of the greater part of the Gulf, in July and August, was 32.4 or more. Off Cape Cod the curve for this value lies about twenty miles off shore; but abreast of Cape Ann it swings eastward toward Cashe's Ledge, cor- responding to the intrusion of fresh water in that region. It then curves toward the coast once more, enclosing Piatt's Bank, whence it runs northeastward almost to Monhegan Island, enclosing Stations 21 and 26. Off the mouth of Penobscot Bay, as already noted, it is forced ^ bulletin: museum of comparative zoology. far off shore (Plate 3) ; but it then approaches the coast once more, water of this or higher sahnity washing the outer islands from Mt. -Desert to the Grand Manan Channel. The salinity of the whole of the Gulf to the south and east of this curve was probably above 32.4; but we have no data on the salinity in the head of the Bay of Fundy. It is probable that the curve of 32.6 enclosed Cashe's Bank, where the violent tides must cause an active vertical mixing of water, and the Grampus crossed it about half-way between Stations 25 and 27, whence it runs in a direct line northeastward, coming close to the coast at Moose Peak. But the water in the Grand Manan Channel was not so salt as this. Whether or not this curve entered the Bay of Fundy is not known; nor can we absolutely establish the occur- rence of water with salinities between 32.4 and 32.6 along the west coast of Nova Scotia; but the facts that water only slightly more saline was found at Station 29 on German Bank, and that there is a considerable discharge of fresh water from the numerous small rivers along this coast suggest that the coast water was fresher than 32.6. Surface salinities above 32.6 were encountered generally over the eastern arm of the deep basin, the value at Station 27 being 32.6; Station 28, 32.75; Station 29, 32.7; and Station 31, 32.84. Unfortu- nately no sample was collected at Station 30. Salinity at intermediate depths. — The table of salinities (p. 139) shows that in no case was the water sal test on the surface; while at most of the stations there was a rapid increase in salinity from the surface downward, though the rate varied in different localities, as shown by the sections (fig. 21-28). At five Stations, 2, 7, 11, 27, 43, samples were taken at three or more levels, thus allowing a satisfactory plotting of curves for the mouth of Massachusetts Bay, off Cape Cod, the western and eastern arms of the 100-fathoni basin, and the trench west of Jeffrey's Ledge. x\t the other stations only surface and bottom salinities are known; consequently the curves are only approximate. But inasmuch as the known curves are all practically parallel down to fifty fathoms or so, they give a guide for reconstructing the others. The type of curve is strikingly different from the temperature curves, being regular and gradual, without the sudden dislocations which characterize the latter, though the increase in salinity is usually most rapid between the surface and fifty fathoms. They show, furthermore, that over the deeper parts of the Gulf the increase in salinity noted on the surface as we go eastward from Cape Ann, extended to the intermediate depths and to the bottom as well. Thus, taking successively Stations 7, 23, 27, 28, (figs. 21, 22) the curves show BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 65 that each was salter than its predecessor at all depths. The bottom salinity at Station 24 is an apparent exception to this generalization; but this Station, like Station 2, lies in a circumscribed trough of the sea bottom, and it is probable that the salinity at the level of the en- closing sill, eighty fathoms, was almost as high as it was at the bottom, just as it was at Station 2. At Stations 8, 9 and 16 (fig. 23, 26) i. c. 0 5 10 15 20 25 50 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 no 115 120 125 1.6 ,< ' i ( }. ) 31 .1 .2 7 ^ - 5 ,6 .7 .S .s 33 .1 .1 3 4 i ) .7 . ! S 1 ^ V \ \ *" \ \ -^ \ \ \ \ V \ \ \ \ \ \ \ ^ ^ ^ \ \ "<: \ \ \ \ X ^ \ \ \ \ J \ A s> \ s\ \ \ N] \ \ I \ s. \ \ N, \. 2 \ s. \ \ \ 1 \\ \ \ \^ T \\ N, \1 \ 43 \ \ \ \ \ \ 23^ 24 \ 7 Fig. 21. — Salinity sections at Stations 2, 7, 23, 24, 43. in the coastal band of low surface salinity, the rate of increase with depth was much more rapid than at the oiT-shore stations, which shows of course that the effect of the fresh drainage from the land is greatest at the surface; and the same is true of Stations 25 and 38 off the mouth of Penobscot Bay. The curves on the off-shore banks, Piatt's, Jeffrey's and German, and in the Grand Manan Channel (fig. 24) are especially 66 bulletin: museum of comparative zoology. important, because of the peculiar temperature conditions which characterized the last two. German Bank and Piatt's Bank bear the same relation to each other in salinity that they do in temperature, the former being colder and salter at the surface, warmer and fresher at the bottom, than the latter. But while Jeffrey's Bank was inter- V 5 10 15 20 25 30 35 40 45 50 55 60 '65 70 75 80 85 90 95 100 105 no 115 120 125 15. 6 ; r ,8 .! 3 35 . .2 .3 i f- .5 . 5 7 .8 .9 54 . .2 .3 ^ \- .5 . S .7 .8 .9 V V \ ■» " \ X '^ \ \ V V \ ^ \ \ \ ''. V s \ V* \ \ \ \, ^ \ s ■^ '\ \ \ N s \'; \ V\ *^ \ A \ \ w \ V \ \ V \ ' \ \3I \ \ \ \ \ \ \ 1 52 \ \ 27 \ * \ 2,8 \ Fig. 22. — Salinity sections at when the surface and bottom Stations 27, 28, 31, 32. Curves are dotted salinities alone are known. mediate between the two in temperature, it had a lower salinity at all ■depths than Piatt's Bank and it was fresher down to about thirty fathoms than German Bank, showing the influence of fresh water from the Penobscot. The increase in salinity with depth was very slight on German Bank, the difference between surface and bottom BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 67 5 10 15 20 12 .. 5 A V \ ) .( ) .< ' i 5 .^ ) i 2 . i » •) "■ -. ^ -*_ .. Orrj I. '- .!; ^ •-. -■-. ■"•■ '-, ..^ -.. S 16 Fig. 23. 5 10 15 20 25 30 55 40 45 50 55 21 : ' 3 4 .5 .6 7 .8 C 33 1 2 .3 i 5 ^•l ■' '~ ,\ "i '< * — ^ 29 \ ', 1; Ni '\ \ *4 '* \ '\ 35 25 ^ 23 Fig. 24. Fig. 23. — Salinity sections in the coast water, Stations 8, 16, and Orr's Island. Fig. 24. — Salinity sections on Piatt's Bank (Station 25), Jeffrey's Bank, (Station 23); German Bank (Station 29), and in the Grand Manan Channel (Station 3.5). 5 10 15 20 25 30 35 40 45 50 55 (SO 65 70 ?5 80 85 90 95 100 2.3.^ \ .E ; .6 . 7 .8 .S > 33 . .2 .3 .4 c .£ ) .7 .S .9 3 4- . .2 ■3 ) .4 .E ) .( ) .1 ' .{ ^ . \, V \ ■% ^ ~\ '\ ~\, \ \^ ■ \ ~~\ \ ~\ 33 \ \ \ 36' 38 '\ \ J Fig. 25. — Salinity sections at Stations 33, 36, and 38. 68 bulletin: museum of comparative zoology. being only .2%o; and in the Grand Manan Channel (Station 33) there was virtually no difference in salinity at different depths, i. c, we find a reproduction of the temperature curve, though at all depths it was somewhat fresher than the water over German Bank. Salinity at twenty-five fathoms (Plate 2). — We have only a few samples at precisely this depth; but the salinity sections, and samples taken at several stations a little deeper or a little shallower than Fc.^ZZJt 4 • .5 i ) .? f r \ 5 52. .1 C 2 ' 4 ^ .5 ■ .i J 7 .? \ S 1 33 . r ./ i 2 i A ^ .! ) .6 .7 5 10 15 ao 25 20 55 m 45 50 55 60 65 70 75 80 85 90 95 100 '~-. ^ •^ ^ ^- \ "'- ^» -«, X ~N \ ■'», .^ ^< nV \ \ ^^, ';v V \ \ 15 ■'■s^' 'x^ N 's ^-, \ X \ '\ 9^ \ \\ \I9 \ *\ ',2 1 \ \ II k \ Fig. 26. — Salinity sections at Stations 9, 11, 15, 19, 21, from Ipswich Bay to Monhegan. twentj'-five fathoms, afford sufficient data for tentative mapping of the curves for the various values at this level. It must be remembered, however, that it, and the following charts, are not offered as final. At twenty-five fathoms the salinity for the whole of Massachusetts Bay, and for an area extending eastward some thirty miles over part of the deep basin was between 32.5 and 32.6. A^d comparison with the chart of surface salinity (Plate 1) shows that the curve of 32.6 in this region reproduces the eastward swing of the curves of 31.8 and BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 69 32.4 on the surface. North of Cape Ann there was a band of compar- atively fresh water, of 32.2 to 32.3, washing the coast along the twenty-five fathom curve, extending northeastward as far as Monhegan Island, some ten miles broad, /. c, roughly corresponding to the fresh coast water noted on the surface in this same region. But it didnot pass around Cape Ann into Massachusetts Bay. The band of water with salinities between 32.4 and 32. G, which was from thirty to 'fifty ¥ 5 10 15 20 25 30 55 Fig. 27. — .Saliaity sections in IMassachasetts Bay, Stations 1, o, 6. 5. 6 7 i \ S ) 32 .1 .2 2 i 4 c ) .6 7 i 5 .? 1 55 *-, *'' ,'\ ~~~. ^ '■"', ^>, 6 5~ ,1'^ ■/n ~-:: V *> '{« 15 € ) .7 ' .8 -^ 32 . .*; . .3 4 \- .1 ) .e ) < ' i i c ) 53 -.^ ~^ ^ 1 ,^44■ "v -^ 45 N, '~- ' \. 46 ^ ^^^ ^^v^ N -> > v» ^ _j __ 5 10 15 20 25 30 55 40 45 Fig. 28. — Salinity sections in MassacliiLsetts Bay, Stations 44, 45,' 46. miles broad abreast of Massachusetts Bay, became very narrow north of Cape Ann, the two curves lying close together as far as Monhegan. Beyond this point, i. e., in the mouth of Penobscot Bay, we have no data on the coast water from depths as great as twenty-five fathoms, or until Petit Manan is reached. But at Station 33 the twenty-five fathom salinity was 32.68, and judging from temperature and tidal currents, it is probable that the curve of 32.6 followed the twenty-five 70 bulletin: museum of comparative zoology. fathom curve from JNIt. Desert Island to the southwestern end of the Grand Manan Channel. Over the central part of the Gulf, including Piatt's Bank as well as most of the 100-fathom basin, the salinity at twenty-five fathoms was above 33%. But the curve for that value runs off shore far enough to exclude the whole of Jeffrey's Bank, thus suggesting the southerly swing of the curve of 32.4 on the surface, though not exactly duplicat- ing it. It then turns northward toward the coast, including in its sweep the whole of the eastern branch of the deep basin, as well as part of the coastal bank off Nova Scotia. The saltest water found at this depth was not at Station 31, as was the case on the surface, but at Station 28 (33.4). In spite of this discrepancy, however, the twenty-five fathom level corresponds to the surface in the presence of intrusions of comparatively fresh water off Cape Ann and off the mouth of the Penobscot, and in the fact that the saltest water was over the eastern edge of the 100-fathom basin. Saliniti/ at fifty fat Jioms. — The curves at fifty fathoms (Plate 3) show the same influx of fresh water off Massachusetts Bay as do the charts for twenty-five fathoms and for the surface, though to a less degree. But I must point out that the charts for the different levels are not strictly comparable with one another in this region, because almost the whole of Massachusetts Bay, as well as the long ridge formed by Jeffrey's Ledge, running some forty-five miles northeasterly from Cape Ann, is shallower than fifty fathoms. The salinity over most of the Gulf at this level was above 33. But along the shore from Cape Ann northward, in the trough west of Jeffrey's Ledge and as far as Monhe- gan, the salinity was lower, between 32.8 and 33. ; and in the isolated basin in the mouth of Massachusetts Bay (Station 2), the salinity at fiftv fathoms was 32.8. The curves at this level hardlv show the southerly swing off the mouth of Penobscot Bay so pronounced at higher levels, the curve for 33 running parallel to the coast, along the fifty fathom line, from Matinicus Rock eastward. This fact, of course, shows that the fresh water from the Penobscot had little or no influence at this depth, although its presence was evident nearer the surface. Over the eastern branch of the 100 fathom basin the fifty fathom salin- ity was above 33.6, the highest being 33.8 at Station 28, while at Station 31, so salt at the surface, the fifty fathom reading was only 33.5. The lowest salinity at this depth was in the Grand Manan Channel, 32.65, practically the same as at the surface. Bottom salinity. — ^ The bottom salinities of the Gulf of ]Maine (Plate 3) are largely dependent on depth, for, as we have seen, there BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 71 was a steady rise in salinity from the surface downward, at all our stations except in the Grand Manan Channel and on German Bank. The bottom salinity below the 100-fathom curve varied from 33.5 to 34.54. There is little if any evidence that the wedge of fresh water abreast of JNIassachusetts Bay, so noticeable from the surface down to fifty fathoms, influenced the bottom water, for the salinity curves at the bottom show very little easterly swing in this region, and that little is probably the result of the bottom contour. The same is true also of the influx from the Penobscot, because the southerly swing of the curve of 33.8 agrees with the bottom contour, following the slope at about the hundred fathom line. It likewise crosses the mouth of IMassachusetts Bay at one hundred and twenty fathoms, rising to about eighty-five fathoms off the northern end of Cape Cod. But it does not enter the trough west of Jeffrey's Ledge, for here the salinity of the bottom water in sixty to eighty fathoms is only 33 to 33.2. North- eastward from Jeffrey's Bank the 33 curve rises higher and higher on the coastal slope until finally water of this salinity was found at about fifty-five fathoms off Petit Manan. The curve must then turn off- shore, for the bottom water in the Grand Manan Channel was only about 32.5-32.6. No station was made on Grand Manan Bank; but judging from conditions on the other banks, it is not likely that the bottom water had a salinity as high as 33. The same is also true of Lurcher Shoal. On German Bank, also, the bottom water was fresher, only 32.9 in thirty-five fathoms; hence it is probable that the 32.6 curve came close to the surface along the west coast of Nova Scotia. The bottom salinity of Piatt's Bank was above 32.5; and it is probable that this was the case on Cashe's Ledge likewise. On the other hand the circumscribed deep basin in the mouth of Massa- chusetts Bay (Station 2) had a considerably lower bottom salinity, 32.92, than the waters at corresponding depths further east. Over the eastern arm of the 100-fathom l)asin the bottom salinity was 34 or over, the highest values being at Station 28, 34.5; Station 32, 34.1; and Station 36, 34.3, in one hundred and twenty, ninety, and one hundred fathoms respectively. But at Station 27, only a few miles west of the saltest spot, the bottom salinity at 100 fathoms was only 33.9. SalinUij profiles. — The profiles (fig. 29-33) can not pretend to as great accuracy as those for temperature, because the number of observations is much smaller; and they are necessarily largely recon- structed from the salinity sections. But if regarded only as prelimi- nary, they are useful as showing general distribution of salinity. 72 bulletin: museum of compakative zoology. o 'S o M m w 1-5 00 a a a o o a CI o -*^ ai o m o a (D ft 03 a I « p4'=' O O oooooooooooo BIGELOW: EXPLORATIONS IX THE GULF OF MAINE. 73 The profile from Boston Light-ship to German Bank (fig. 29) shows conditions in Massachusetts Bay, over both arms of the deep basin, on Jeffrey's Bank, and over the coast slope of Nova Scotia. From Station 6 to 7 salinities were very uniform at all depths, except for a slight upwclling of salt water above twenty fathoms at the westerly end, thus paralleling the temperature profile (fig. 16) and for the fact that there was no appreciable increase in salinity below forty fathoms in the isolated basin at Station 2, i. e. below the depth to which the enclosing sill rises. Passing easterly from Station 7, the entire mass of water above seventy fathoms becomes salter, all the curves ap- proaching the surface, that for 32.5 rising from thirty fathoms to the surface, while the curve for 33 lies at about twenty-five fathoms at Station 24, instead of at fifty fathoms as at Station 7. Below seventy fathoms, however, there is very little difference between the two sta- tions. Our profile thus shows that the wedge of comparatively fresh water off ^Massachusetts Bay was not traceable below about seventy fathoms. Over Jeffrey's Bank the water was appreciably fresher at all depths than it was either west or east of it, the curve for 33 showing a pro- nounced downward swing from twenty-five fathoms at Stations 23 and 24 to fifty fathoms at Station 25. But its upper twenty fathoms, though fresher than at Station 24, had a higher salinity than in the region west of Station 7. The whole of the eastern basin was Salter at all depths than the regions west of it, the curve of 33.6 rising to within about forty fathoms of the surface at Station 28, whereas in the western basin, water of this salinity was only found below ninety- five fathoms. At all depths down to about twenty-five fathoms salin- ities were highest at Station 31; but below that depth at Station 28; for example, the curve of 33.8 lies at sixty-five fathoms at Station 31, at fifty fathoms at Station 28; and the curve for 34 must show an even more pronounced rise, for water of that salinity or over was found at Station 28, from eighty fathoms down to 120 fathoms, whereas at Stations 27 and 31 the bottom water was only 33.9 and 33.8 respec- tively in 100 and in seventy-five fathoms. Over German Bank, as already pointed out, the water was between 32.7 and 32.9 from surface to bottom. The west to east extent of the fresh Penobscot water is shown by a profile running from Piatt's Bank across JeftVey's Bank to the neighbor- hood of Mt. Desert Rock (fig. 30) and the breadth of the coast-band of comparatively fresh water off the mouth of Casco Bay is illustrated by a profile from Station 15 to Station 24 (fig. 31). A similar profile 74 bulletin: museum of comparative zoology. Fatk. 25 Fig. 30. — Salinity profile from Piatt's Bank (Station 23) toward Mt. Desert Rock (Station 32) crossing Jeffrey's Bank (Station 25). Fig. 31. — SaUnity profile from the mouth of Casco Bay to Station 24. BIGELOW: EXPLOKATIONS IN THE GULF OF MAINE. 75 running southeastward from Mt. Desert (Fig. 32) shows the increase in salinity passing off shore in that region. In Massachusetts Bay two pairs of Stations, 6 and 45, and 5 and 4(i, (figs. 27, 28) were taken six weeks apart, purposely to show seasonal change, if anv. But the Fa. n 51 J^8 0 sections show that at both 46 and 5 the salinity at the surface was 31.6, at 30 fathoms 32.5, and also that there was apparently noth- ing to separate Station 6 from Station 45, at both of which the surface sa- linity was 31.9; though as the depth at the former was twenty-five and at the latter forty fathoms, only two samples being taken at each, it is possible that there may be some slight divergence in the interme- diate zone. In short, these four stations certainly do not suggest that there was any seasonal change in the salinity in Massachusetts Bay during our absence, although there was a very pronounced rise in temperature (p. 58) at all depths below five fathoms. Stations 44, 45, 46, all taken on the same day, afford a profile across the Bay, from south to north (fig. 33). The curves show that the core of fresh surface water was thickest in the northern half of the Bay. And as Station 6 is, as we have just seen, interchangeable with 45, and 5 with 46, it is clear that this is the characteristic condition in mid- summer. In the southern half of the Bay, the curve of 32.2, found at twenty fathoms at Station 46, rose to within eight fathoms of the sur- face; and the surface salinity was 32 instead of 31.9. But below twenty fathoms the salinities were slightly lower at Station 44 than in the centre of the Bay. Thus we find reproduced, but on a much smaller scale, the spreading of the salinity curves so pronounced on German Bank, and in the mouth of the Grand Manan Channel. And as pointed out (p. 56) the same thing was true of the temperatures. Fig. 32. — Salinity profile from near Mt. Desert (Station 37) to Station 28. 76 bulletin: museum of compakative zoology. In the northerly end of the profile Station 1 is introduced, to show- how the cold salt bottom water wells up close to the shore. However, as^pointed out in the discussion of temperatures, this phenomenon is Fig. 33. — Salinity profile across Massachusetts Bay, August 13 (Stations 44, 45, 46). sporadic, probably the result of offshore winds driving the surface water away from the coast, their place being taken by water from below. Conditions at Station 1 show that the effect may be felt to as great a depth as 20 fathoms. Density. The three features of sea w^ater most interesting to the oceanog- rapher are temperature, salinity, and density; the former because of its biological importance; the second because it is the only safe clue to the geographic origin of water-masses; and the third because of its importance as determining circulation, both vertical and hori- zontal. The last is a product of the first two and of a third factor, namely pressure. And we must never lose sight of the fact that as it is determined by temperature as w^ell as by salinity, it is a temporary quality, changing as the water becomes colder or warmer. In the accompanying table (p. 141), the densities in situ are calculated from BIGELOW: EXPLORATIONS IX THE GULF OF MAINE. 77 Knudsen's (1901) tables and from Ekinan's (1910) tables of sea water under pressure. Such calculations are approximately correct arith- matically, but notice must be called to the fact that the probable limits of error are the sum of the two observational errors, first for salinity, i. e., =*= .02 of salinity (p. 40), second for temperature, which is =fc .3° F, approximately .15° C. Now the sum of these errors has a considerable efTect on the calculated densities, and for this reason the fifth decimal point is disregarded in the table. Of course a much higher degree of accuracy could be, and is, obtained with improved instruments, for example, during the North Atlantic cruise of the Michael Sars in 1910 (:\Iurray and Hjort, 1912). But it would be misleading to claim better results with our instruments. R.25 .2 4- .6 i 5 24 .2 4 1- .6 .< 3 25 .2 4 ; 5 .8 26 .2 4 - .( 3 .8 u 5 10 15 20 25 50 55 40 45 50 55 '- ^ \ ■*x. \ *^ \ \ " N.. \ '. ^ \ *^ '-. \ \ V •v \ 29 , \^ ■~-, 1 23 ■','25 Fig. 34. — Curve of density ui situ at Piatt's Bank (Station 23); Jeffrey's Bank (Station 25) and German Bank (Station 29). The correction for pressure has often been disregarded, especially in shallow water; but it can easily be applied from Ekman's tables. In depths less than fifty fathoms it is of little practical importance, but by the time 100 fathoms is reached it is by no means negligible. For example, at Station 28, 120 fathoms, the density at the tempera- ture in situ without pressure correction, is 27.02; with pressure correc- tion, 28.03. In the accompam-ing table the pressure correction for depths less than fifty fathoms is calculated by the use of Ekman's table IV alone, which is sufficiently accurate for our present purpose. The most important thing which the table and curves (fig. 3-1-36) show is that there was a steadj^ increase of density at every station from the surface down to the bottom, which, as we now know, is the normal 78 bulletin: museum of COMPARAXn^E ZOOLOGY. condition in all ocean waters during the warm season, though there are temporal and local inversions due to temperature conditions in winter (Helland-Hansen and Xansen, 1909). But the rate of increase varies greatly in different regions, there being two very different tx'pes of vertical distribution in the Gulf. The first, exemplified over Jeffrey's and German Bank, and in the Grand Manan Channel, Station 25 (fig. 34), Station 29 (fig. 34), Station 35 (fig. 36), shows only a very slight increase from surface to bottom ; but in the second, comprising practically all the other stations, there is a large rise, with slowly decreasing rate from the surface downward. The curves for fJ2€J S .25.5 A I ,6 > .i i 24 .5 4 I .i ) .8 25 .5 ! i 1- .G X M^ X I \ .6 i \ 2 u 5 10 15 20 25 50 55 40 45 50 55 60 65 70 75 ^ -^ ^ ^ \ ^ ii [^ ^^^ 26 2 \ '^ *x ^ -^^ <^ ^ ^ \ N, \ \ X \ \ \ \ \ \ V \ s ■ \ s ^ 80 Fig. 35. — Curves of density in situ at Stations 2. 11. 26. the last recall the salinity curves at corresponding stations; but the difference between surface and bottom was in every case considerably greater in the former than in the latter, at corresponding stations. The most important conclusion to be drawn from the density curves is that over the whole deep basin, in Massachusetts Bay, and along the coast from Cape Ann to the Penobscot, the water was in very stable vertical equilibrium during July and August; but that on Jeffrey's and German Banks and in the Grand IManan Channel the difference in density in different depths was so slight that it wouldF offer very little resistance to vertical circulation. In comparing the BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 79 densities for Stations 6 and 5 with those of 45 and 46 it is evident that vertical stabiHty in ]Massachusetts Bay decreases with the advance of the season; pointing to the inversion which no doubt takes place there in winter. A profile from Boston (Station 6) northeastward to German Bank, via Jeffrey's Bank (Station 25), and Station 31 (fig. 37) shows the 5 10 15 20 25 50 55 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 ?8 25 ' 2 4 . 5 . 3 24.- I 4 .« S . 3 25 .; I 4 ■ S . 3 26 .< ? 4 .< 5 . 3 27 .. I 4 .( 5 . 3 28 ^ ■--, 1 4^ \ •^ <^ \ 7 ■~~- ^ s- \ s V 5 ^ ~^ k ■^ N N \^ ::n \. \ V ■i: \ \ \ ^ ^ \ \ \ N \ \ \ h^ \ \ \\ \ \ \\ \ \ \ k \ \ \ K \ \ \ \ \ ^^ A \ \\ w \^ \ \A \ ^ \ \ \ 115 120 125 150 \ \ \ Fig. 36. — Curves of density in situ at .Stations 7, 27, 35, 43. relative distribution of lighter and heavier water over the northern part of the Gulf. Above the two deep basins heavy water, distin- ^guished arbitrarily by the curve of .020, rose close to the surface, W'hereas over the two banks the whole column of water was of lower den- 80 bulletin: museum of comparative zoology. (^O b£ ^ IV5 BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 81 sity than this. But the water of the eastern basin was appreciably denser at all depths than that of the western, corresponding densities being about ten fathoms deeper in the latter than in the former. And corresponding densities were found twenty to thirty fathoms higher over German Bank than over Jeffrey's Bank. The profile shows also in a graphic way how much more rapid the downward increase was over the basins than over the banks, and consequently how much more stable, vertically, their waters must have been. At the western end of the profile there was again an increase in density as compared with the western half of the basin, a phenomenon consequent on the up- welling of cold, salt bottom water in this region, while at Stations 2 and 7 density near the surface was very low, corresponding to low surface salinity. In the trough between Jeffrey's Ledge and the coast the density agreed closely with that of the western basin, except that it was rather higher on the surface, corresponding to the low surface temperatures of this region. And passing northeastward along the coast we find the vertical range progressively less and less, until in the Grand Manan Channel the difference between surface and bottom was only .6 at 45 fathoms. The information our cruise afforded as to densitv is insufficient even for the northern half of the Gulf, but so far as it goes, it shows that two distinct water masses can be distin- guished, a light over the western, a heavy over the eastern basin, partially separated by the disturbed conditions caused over Jeffrey's Bank by the influx of fresh water from the Penobscot. Color. The color of the sea is of minor importance in oceanography: but it can not be neglected, because it helps to form the physical complex, in which the plankton finds its biological environment. The color, described by percentages of yellow as indicated by the Forel scale, is given in the table (p. 82). At the off-shore stations it varied from 27% (Station 43) to 14% (Stations 7 and 23), usually being 20%: in Massachusetts Bay it was 20% at all stations at which a record was made (Stations 2, 4, 6, 44, 45). West of Jeffrey's Ledge the color was 14% at Station 9; but grew greener as we went north, being 20% at Station 11, 27% at Stations 13 and 14. Off the mouth of Casco Bay it was 27%, inside the Bay 27% and 35%. Over the northeastern part of the Gulf as a whole, the color was 20% yellow, except close to the shore (Stations 33, 37) where it was 35%. This distribution of waters of different colors does not correspond either to temperature 82 bulletin: museum of compakative zoology. or to salinity, for the bluest water was not the saltest, while the coldest water was neither bluest nor greenest. The plankton may give the necessary clue. Color, in % of Yellow in the Forel scale. Sta. Color Sta. Sta. Color Sta. Color Sta. Colo 2 20 13 27 21» 27 29 20 39 20 4 20 14 27 22 27 31 • 20 40 20 6 20 15 27 23 14 32 27 41 20 7 14 16 27 25 20 33 35 43 27 8 20 17 35 26 20 35 20 44 20 9 14 Orr's I. 44 26* 20 36 20 45 20 10 20 19 20 27 20 37 35 11 20 21 27 28 20 38 20 The color of the Gulf of Maine agrees fairly well with that of the southern part of the North Sea, with the English Channel, and with the coast water of the Bay of Biscay (Schott, 1902, pi. 36). Up to the present time we have no records of the color of the water along the coast of the United States from Cape Cod south, or for the Gulf of St. Lawrence. Transparency. Measurements of transparency were taken with the disc (p. 00) at eighteen stations. In the clearest water (Station 23) it was visible at 8.2 fathoms ; but it usually disappeared at from four to five fathoms. There was little, if any, correlation between color and transparency at these stations, for though the water was most transparent where bluest (Station 23), it was not least so where greenest, but where the percentage of yellow was only 20% (Station 38). Transparency, in fathoms. Sta. Trans. Sta. Trans 4 3.5 31 4 11 6 36 4 12b 6 37 4 14 6 38 3 15 4.5 39 4 16 3.5 40 6 22 7.2 41 5 23 8.2 43 5 25 6.5 44 5 BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 83 Circulation in the Gulf of Maine. Circulation in the Gulf may be expected to be of three types: — 1, tidal, which is proverbially violent in the northeastern part of this region; 2, the slower but more constant vertical or horizontal move- ment of water resulting from different density gradients at different regions, or from the presence of an actual ocean current, if there be one; and 3, sporadic movements of the water, due to prolonged or violent winds. Tidal currents. — A considerable number of measurements of tidal movements have been made on the surface of the Gulf of Maine by the U. S. Coast and Geodetic Survey, by the British Admiralty, and by the Tidal Survey of Canada in charge of Dr. G. B. Dawson; but so far as I can learn, the only accurate records of bottom currents are the few taken on the Grampus last summer. The earlier surface records, for off shore stations, are limited to the east coast of Cape Cod, Stell- wagen Bank and the channels north and south of it, George's Shoal, the Eastern Channel, Brown's Bank, the west coast of Nova Scotia, and the Bay of Fundy; these the 1912 cruise of the Grampus extends to the central part of the Gulf and to the coastal region between Cape Ann and the mouth of the Penobscot. Although our records are too few for a complete survey even at a given station, we always at- tempted to take them as close to the mid-period of flood or ebb as possible, so as to obtain the mean direction and velocity of the cur- rent for a given tide; but of course, this result could be expected only in regions where the current was fairly constant for the major part of each tide. The sum of all available observations suggests that the violent surface currents of the Gulf, noticed by every navigator, are purely tidal, the mean flow of ebb and flood being in general about equally strong at a given locality; but the mean directions of the two are not always pre- cisely opposite. The general rule is that " along the whole line between Nantucket shoal and Cape Sable Bank the ebb current runs south- wardly. . . .the flood current northwardly. ..." (U. S. Coast Pilot), and along this whole line the currents are swift (1.1 to 1.6 knots at their height) . The tidal wave divides over the basin south of Jeffrey's Bank and Cashe's Ledge, the flood currents west of here turning west- ward toward Massachusetts Bay, and toward the coast between Cape Ann and Portland. Abreast of Casco Bay (Stations 14, 40) the flood flows nearlv due north; but east of Jeffrey's Bank the general direction 84 bulletin: museum of comparative zoology! of the flood near shore, is N. N. E. toward the Grand Manan Channel and the Bay of Fundy, and along this coast the velocity increases steadily from west to east, the rate in the channel being two knots. Along the west coast of Nova Scotia the mean direction of the flood- current is nearly north. The flood is weakest in the northern part of Massachusetts Bay, and along shore from Cape Ann to Portland, as shown in the Table (p. 143), though there are strong tidal currents off the mouths of large rivers, and tide rips ofl" Portsmouth (Station 11). In the central part of the Gulf (Stations 7, 27) the current is about .5 knot; but along the Nova Scotian Coast and off the mouth of the Bay of Fundy it occasionally attains velocities of more than two knots, with extensive and dangerous tide-rips on the various shoals, for example the Grand Manan Bank. In a general way the ebb is the reverse of the flood, flowing out of the Bay of Fundy in a generally S. W. to S. S. W. direction, and around the coast of Nova Scotia to the S. and S. E. Along the coast of Maine from the Grand Manan Channel to Mt. Desert the ebb flows about S. W. But the current in the central part of the Gulf is about S. by E. Off Casco Bay the ebb is southerly; along the coast from Portland to Cape Ann it sets in general toward the E. S. E. but there are various local currents here, yet to be explained. The strength of the ebb current is proportional to that of the flood, strongest off the mouth of the Bay of Fundy and along the coast of Nova Scotia; progressively weaker to the westward. The data is insufficient to show whether the tidal currents result in any definite eddy movement of the waters of the Gulf, nor have I been able to find in them any evidence of an inflow, or alongshore flow within the Gulf, such as might be credited to a branch of any constant ocean current. This question was thoroughly studied by Dawson (1910) for the Bay of Fundy and for the Nova Scotian Coast, between the mouth of the St. John and Cape Sable, in 1904 and 1907. And his general conclusion is that ebb and flood are almost opposite, veering at slack water onl}-, if at all; and that there is little indication of any movement of water in a dominant direction. The mean compass-bearings and strengths of the currents on Brown's and George's Banks as given on the U. S. Coast Survey charts suggest a drift from northeast to southwest. But the data on the tidal cur- rents of George's Bank given by Mitchell (Rept. U. S. Coast and Geo- detic Survey, 18S1, p. 175) show that there is a slight easterly drift, and it is so represented on the current chart in the coast pilot. (U. S. Coast Pilot, part 3, 1912, chart facing p. 9). And although most of BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 85 the current charts which have appeared show a southwest flow along the outer edge of George's Bank, next the Gulf Stream, it is a question whether this flow is a contant, or even a dominant one. Surface and bottom currents. — To obtain a satisfactory knowledge of the tidal currents at any locality, it is necessary to make observa- tions at intervals throughout a twelve-hour period, to insure read- ings for both flood and ebb, because the time of turning of the tide at the bottom often difters by a considerable period from the time of slack water on the surface. Nevertheless, our few isolated observa- tions are worth passing notice because they are the first attempts to measure the bottom currents of the Gulf of ]Maine with modern instru- ments. The diagrams (fig. 38) illustrate the considerable strength of the bottom currents even in the western side of the Gulf; and in the northeastern part, for example over German Bank, they are even stronger. The only region where enough observations were taken to allow a tentative statement of the relations of bottom to surface currents is the northern half of ^Massachusetts Bay (Stations 1, 2, 4, 5, 6). The surface current flows into this part of the Bay toward northwest and west at the height of the flood (Station 5) turning at least one half hour before the time of high water at Gloucester, and flowing easterly during the first half of the ebb (Stations 1 and 4). We made no records for the last three hours of the ebb. A few miles further off shore the direction at mid-ebb was southeast (Station 2) ; and in the centre of the Bay (Station 6) N. N. E. (all bearings being magnetic). In the southern half of the Bay the flood current ran toward the southwest, the ebb toward the northeast. These observations are not sufficiently extensive to show whether or not there is any dominant drift along- shore. But tidal records taken by the U. S. Coast Survey at the mouth of the Bay suggest that it may be occupied by an eddy-like circulation flowing slowly from north to south, there being a decided drift to the northwest near Cape Ann, with an easterly movement on Stellwagen bank and near Race Point (U. S. Coast Pilot, part 1 and 2, 1911, p. 151). The bottom currents in Massachusetts Bay dift'er very noticeably from the surface ones (fig. 38) not only in being as a rule weaker, but in flowing in a different direction. At all the stations in the central and northern part of the Bay, the bottom flow was easterly, the records being made a few minutes before high water (Station 1), tv\o-hour ebb (Station 4), mid-ebb (Station G), and early flood (Station 5). This data, so far as it goes, suggests that if there be any tidal flow to the west on the bottom it must be restricted to 86 bulletin: museum of comparative zoology. the last two hours of the flood, veering again to the eastward shortly before high water. There was no bottom current at Station 2, and inasmuch as that station was occupied at the mid-ebb, when the bottom current further N S \ N s. S 4 N S 5 N? S 6 N S 7 N X s 8 N s 9 N J T.. a n -, "^ f \ « U- s II N 14 N s 25 N N. s 27 N N s 38 N s 45 Pig. 38. — Currents, Stations 1. 4. 5, 6, 7. 8, 9, 11, 14, 25, 27, 38, 43. Surface current > ; bottom current > . The strength of the current is represented by the length of the arrows, 3 cm. = 1 knot per hour. within the Bay attains considerable velocity, its absence is no doubt to be explained by the fact that the bottom here is an isolated pocket somcjtwenty fathoms deeper than its enclosmg sill, with a consequent BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 87 separation from the general bottom circulation. The highest bottom velocity recorded in Massachusetts Bay was .25 knots per hour. The observations at Station 7 show that over the western basin oppo- site Cape Ann the surface current begins to flow westerly with the considerable velocity of .5 knot, at least two hours before it is low water at Cape Ann; but not the bottom current, for the latter, a few minutes later, was still flowing to the east, though slowly. There are four stations between Jeffrey's Ledge and the mainland in depths from twenty-five to eighty fathoms, at which bottom as well as surface readings were taken. Stations 8, 11a, on the flood; Stations 9, lib and c, 14, on the ebb. On the surface the flood current runs to the west (Station 8), orW. by S. (Ha), velocity .3 to .4 knots per hour, and the three sets of observations at Station 11 show that the current was still running to the westward two hours after high water at Cape Ann and Cape Porpoise, though it had veered from west to north by west. But there were several active tide-rips in the vicinity, which were probably responsible for this apparent on-shore flow during the ebb. At Station 14, there was a strong current flowing southwest (.6 knot) two hours before low water at Cape Porpoise, only eight miles distant. But at Station 9 it had started to run slowly to the N. E. by E. one hour after high water. On the bottom the current was easterly in every case (Station 8, 9, 14, lie) except at lib, where there was a very slow movement to the N. N. W. One of these records (Station 8) is at four-hour flood, the others are at various stages of the ebb. At Station lib, two-hour ebb, the bottom flow was toward the N. N. W.; but one half hour later it had veered to the E. by N., i. e., toward the extremity of Jeffrey's Ledge. At Station 14 there was a .3 knot cur- rent on the bottom toward the E. by S. the surface flow being S. W. On Jeffrey's Bank (Station 25) the bottom current was to the E. S. E. almost at right angles to the surface flow (N. by W.), three hours after high water at Portland and Rockland. Our one bottom reading in the eastern basin (Station 27) revealed a quarter-knot current, running southerly, like the surface flow, on the early ebb. Off Cape Cod, our single reading (Station 43) showed that the bottom flow was still toward the northwest, on the early ebb, although the surface current was already flowing to the southeast. Circulation as shown by temperature and salinity. — Since we have seen that the surface currents of the Gulf, though often violent, do not demonstrate the existence of any circulation on broader lines than that caused by the tides, we must turn to salinities, temperatures, and 88 BULLETIN : MUSEUM OF COMPARATR^E ZOOLOGY. densities in the attempt to reconstruct the movements of its waters the most important subject on which the cruise may throw light. Perhaps the most striking oceanographic feature of the Gulf of ^Nlaine in summer, certainly the one which has aroused the most speculation, is the existence of a cold band of surface water which bathes the coast from Portsmouth as far as the Penobscot, and extends thence across the mouth of the Bay of Fundy and along the western coast of Xova Scotia, gradually growing broader and broader to the eastward. If we were to judge from surface temperatures alone we would naturally assume that this cold water was e\-idence of a cold current following the coast ; and it has often been referred to as an Arctic current solely on this ground. But, as we have seen, the surface currents, at least in summer, afford no support to such a view, while serial temperatures and salinities show that the phenomenon can be explamed on very different grounds. The coldest surface water was found over German Bank and in the Grand IManan Channel; but serial temperatures show that this low temperature, at these stations, was solely a surface phenomenon, the bottom waters being much warmer there than at corresponding depths in the basin or on the west coast of the Gulf. Furthermore the mean temperatures for the upper forty fathoms, i. c, for the whole depth at Stations 29, 33, and 35, are no lower than they are in the western part of the Gulf; (Station 29, 49.8°; Station 33, 49.5°; Stations 27 and 2S, 49°; Station 11, 45.7°; Station 7, 49.1°; Station 2, 46.4°: Station 43, 51.1°.) We find, too, that in the northeast part of the Gulf, there is much less change in salinity from surface to bottom than in the western half. And when we take into consideration the extraordinary violence of the tide, both on German Bank and in Grand Manan Channel, and the numerous tide-rips, with which everyone who has sailed these waters is familiar, it can hardly be doubted that the low surface and high bottom temperatures are merely the e\'idence of thorough mixing of surface and bottom waters, caused by the active vertical circulation which necessarily results from the strong currents. Verrill (1873, p. 438) explained the phenomenon correctly when he wrote "the constant mixture of the cold bottom water with the warmer surface waters by means of the strong tides and local wind currents, causes the remarkably low temperatures observed in the shallow waters of these shores." The temperature conditions on JefFrey's Bank re- sult from a similar phenomenon, though as tidal currents are less strong here than they are further to the eastward, the equalization of tem- perature from surface to bottom is less complete; and the diminishing BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 89 range of temperature from surface to bottom at successive stations along the coast from the Penol)scot to Grand Manan, the surface growing warmer, and the bottom colder at least at corresponding depths, is evidently due to the fact that the diminishing force of the tidal currents is less and less effective in causing vertical circulation, so that the waters retain more and more nearly their normal tempera- ture gradient. Exactly the opposite takes place in passing off shore from the mouth of Grand ^Slanan Channel, the temperature and salin- ity range growing progressively greater. The mouth of Casco Bay, i. e., the region where the general trend of the coast changes from northerly to northeasterly, is the dividing line between temperature sections of two types; for whereas the coast waters east of this point were about as much warmer than the off-shore stations on the bottom as they were colder on the surface, the coast water south and west of Portland was no warmer on the bottom than it was off Cape Ann or near Piatt's Bank, though it was constantly several degrees colder on the surface. On the contrary, Station 11, close to the coast, was colder at all depths down to about sixty fathoms than the water east of Jeffrey's Ledge, and the curve at Station 14, off Cape Porpoise, was almost precisely like it, the same temperatures being found from five to ten fathoms nearer the surface at Station 11 than at Stations 23 and 24. Below fifty fathoms the temperatures were about equal. If temperature were the only clue to oceanic circu- lation, we would naturally assume that such a profile indicated an upwelling of cold bottom water. But the salinities of this region, forbid this explanation, because, as the salinity sections show, the in-shore stations were fresher at all depths, whereas, if the surface were cooled by water rising from below, the salinity would necessarily be raised by the same process, and we would expect to find the surface Salter than, or at least as salt as it was at the stations further off shore. But although the temperature readings at Station 11 were lower at all depths down to fifty fathoms, than they were east of Jeffrey's Ledge, the curve for the former wa3 almost precisely the same as it was at Station 2, in the mouth of ]\Iassachusetts Bay a few days earlier, except that the upper ten fathoms were cooler at Station 1 1 ; while the salinity curves (fig. 11) show that the latter was slightly salter than Station 2 at the surface, slightly fresher below thirty fathoms. It is evident that while vertical movements of such a column of water as was met at Station 24 could not reproduce the temperature and salinity conditions found at Station 11, a vertical mixing of the upper fifteen or twentv fathoms of the waters at Station 2 would cause 90 bulletin: museum of comparative zoology. results very similar to the conditions observed over the trench west of Jeffrey's Ledge. And this is probably the correct explanation. Further evidence in its favor is afforded by the fact that diurnal changes of surface temperature are not so great in this region as they are further oft' shore. The profiles show that this mass of coast water is fairly sharply de- fined from the off shore water east of Jeffrey's Ledge in July and Au- gust, by low temperature and low salinity, in which it agrees with the water off the mouth of Massachusetts Bay. And no doubt the con- tour of the bottom is largely responsible for this fact by hindering free circulation of the water below thirty-five fathoms; because although the northern end of the trench is open, and the water there (Station 22) was Salter than it was at Station 11, yet it was so much warmer at all depths that the density, depth for depth, was about the same at the two stations. Consequently there is no dynamic cause for an active flow of water of high salinity into the deep parts of the trench, and the latter retains more nearly the conditions of early summer than does the coast water further north and east. Temperatures and salinities show that the cold bands of water so often observed along the north shore of Massachusetts Bay are evi- dence of upwelling of l^ottom water, probably due to off-shore winds. But in the southern half of Massachusetts Bay, the curves of both these factors, taken with the strong tides of this region, show that the cool surface water is the result of mixing, rather than of upwelling, two forms of vertical circulation which may be perfectly distinct, though they are often combined. The existence of a band of coast water of very much lower salinity than the off-shore water is no doubt the direct result of the vast volume of fresh water poured into the Gulf by the large rivers which empty into it, chief of which are the Merrimac, Saco, Androscoggin, Kennebec, Penobscot, St. Croix and St. Johns, with a combined water-shed of about 45,550 square miles. L^nfortunately we have very little data on the salinities of the Gulf at any season of the year except in mid- summer, but the salinity curves for July and August show that at that season, at least, the fresh river water is localized along the coast, swinging off shore opposite the Penobscot and off Cape Ann. It is true that at that season there is little or no evidence afforded by the surface salinities of an influx of river water in the northeast corner of the Gulf, although it is there that the greatest volume enters, i. e., from the St. Johns and St. Croix. And although this can be partly ex- plained as due to the active vertical circulation in this region, which BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 91 raises the surface salinity by mixing, virtually equalizing the physical properties of the water from surface to bottom, the mean salinities for the upper thirty fathoms show that the water off the Grand Manan Channel is absolutely, as well as apparently, Salter than it is at Station 11, or off Massachusetts Bay; and that Jeffrey's Bank, off the Penob- scot, is intermediate between the two extremes (the figures are: — Stations 33 and 35, 32.5%o; Station 25, 32.6%o; Station 11, 32.3%o; Station 19, 32.4%o; Station 2, 32.2%o). These facts must be ampli- fied by records from other times of year; but so far as they go they point to the conclusion that the coast water flows southwesterly along- shore, with a branch turning southward off the mouth of the Penob- scot; and that it swings eastward as a whole off Cape Ann. The fact that the St. Johns water is less evident, though much greater in amount, than the water from the Penobscot, Kennebec, and Merrimac, can be explained only on the assumption that it is more constantly mixed with salt oft'-shore water than are the latter; an assumption supported by our observation that oceanic salinities are most closely approxi- mated both on the surface and in deeper layers in the eastern part of the Gulf. All this, of course, indicates an in-shore movement of water in this region in August, which mixes with the St. Johns water oft' the mouth of the Bay of Fundy, with consequent changes of salinity; while the occurrence of Salpae over the Eastern Basin is as good evidence, as is the high salinity, that at the time of our visit this oceanic water was an offshoot from the northern edge of the Gulf Stream, not of northern origin. Off the mouth of the Penobscot the salinity curves show that the flow is the reverse, i. e., to the south; but off Casco Bay we once more find a tongue of comparatively salt water approaching the coast, and separating the Penobscot from the Cape Ann fresh wedge. Thus, although the actual movements of tidal currents do not reveal the existence of any general circulation in the Gulf (p. 84), salinity conditions show very clearly that there is an influx of ocean water on the east side of the Gulf; and a longshore movement of the fresh coast water, sending out a southerly tongue off the Penobscot, and swinging eastward off Cape Cod. In other words, the surface of the Gulf as a whole, at the time of our cruise, was probably occupied by two separate eddies, which are reconstructed here from the salinities (Plate 4). The fact that the salinity is lower over the western than over the eastern side of the eastern basin, is due to the eddy, part of the fresh wedge off the Penobscot being drawn into its circulation on the west side. And the comparatively low salinity of the western basin, and the gradual rise of salinity from west to east 92 bulletin: museum of comparative zoology, is similarl}' explained bj^ the indraught into the eddy of the compara- tively fresh water off Cape Ann. Off Cape Cod (Station 43) the water was considerably salter than off Cape Ann, the mean for the upper fifty fathoms being about 32.7, instead of 32.4 as it was at Stations 2 and 7; i. e., in this region the influence of the coast water was felt but Httle. Unfortunatelv we vet hsLve so little data for the salinities of the region south of a line from Cape Ann to Cape Sable, that an attempt to extend the chart of circulation over the southern half of the Gulf would be little better than guess work. Comparison w'ith previous records of tejiperature and salinity. In July and August, 1873, Verrill found nearly the same surface tem- pera tm'C fifteen to twenty miles southeast of Cape Elizabeth as we did last summer, 62°-6o°. Near Seguin Island, August 20, he records 59°; this is very close to our Station 40, where on August 22 our reading was 58°. But by September, 1873, the surface temperature had fallen several degrees below our August records, the surface temperature east of Jeffrey's Ledge, and generally over the western basin opposite Cape Arm being given by Verrill as 57°-58°, i. e., autumn cooling had proba- bly set in by that time. And his records near ■Slonhegan and Matini- cus Rock are from 2°-3° lower than ours a month earlier. But in Massachusetts Bay his records are 59°-64°, suggesting that seasonal cooling in that region was more rapid in 1912 than in 1873. The temperature data obtained by the U. S. Coast Survey in 1874 is of slight value, because the surface readings are not reliable (\ errill, 1875, p. 413, footnote) ; and there is no way of estimating the probable error, which may be several degrees. But so far as they go they sug- gest that the surface of the Gulf was several degrees warmer in that year than in 1912, with surface temperatures of 60-69° in its south- western part. On Cashe's Ledge, and on the northern part of Jeffrey's Ledge readings of 55-58° were obtained, probably an index of active vertical tidal circulation. Dickson's (1901) charts for July and August, 1896, and July and August, 1897, show a very different distribution of surface tempera- tures in the Gulf from what we encountered: for they do not show the cold coast-band east and north of Cape Ann, while in July, 1897, the surface temperature of the whole of the Gulf east of about 69° W. Long, is given as below 59°; the smaller area west of 69° Long., 60° BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 93 or warmer: and in August of that year, he shows all of the Gulf, in- cluding the Bay of Fundy, 59° to 68°, Massachusetts Bay above 68°. But without access to the vessels' logs, and other unpublished data from which these charts were compiled, it is useless to discuss them critically further than to point out that the distribution of tempera- tures within the Gulf represented on them does not accord with condi- tions in 1912, with Verrill's observations, or with the occasional surface readings which I have made in other years. The surface temperatures taken by the Grampus in July, 1908, while crossing the mouth of Massachusetts Bay on her way to the Gulf Stream, (Bigelow, 1909) were 66° to 68°, i. e., appreciably higher than they were in 1912. This fragmentary data suggests that the surface waters in the Bay and over the Gulf as a whole, were colder than usual during the summer of 1912; the result of abnormally low air temperatures during the pre- ceding winter, the coldest in eastern Massachusetts for many years. Unfortunately the only previous records of bottom temperatures within the Gulf, those recorded by Verrill, (1873-1875) are not reliable, as shown by the fact that when two thermometers were used simulta- neously their readings occasionally differed by as much as 4.5°, fre- quently by 1° or 2°; indeed it was the exception that they registered alike, and as Verrill himself pointed out, they rated differently at successive standardizations. His records, taken at their face value, would indicate that the bottom temperatures were distinctly lower in the northeastern part of the Gulf in 1873 than they were in 1912; i. e., the reading, with both thermometers, in 107 fathoms, in Septem- ber, 1873, twenty-three miles southeast of Matinicus Rock, was 39.5°; in 105 fathoms just east of Jeffrey's Bank it was 40°, whereas it was 42.8° at Station 27, in 1912. Fifteen to twenty miles southeast of Cape Elizabeth, the discrepancy is still greater, for Verrill records bottom temperatures of from 36° to 39.5°. But these differ so much from those of the Gr.aj\ipus (41° to about 45°) and are so much lower than he himself records from any other part of the Gulf, that it seems that the instrumental readings were too low. In the deep basin off Grand Manan, Verrill found the bottom temperature 37.5° in 106 fathoms in 1872, but we have no data to compare with his; and this basin is isolated from the exterior by a sill over which there is only about eighty fathoms of water. Fifteen miles southeast of Boon Island, in the trench west of Jeffrey's Ledge, the older record is about 39° (37.5° and 40.5°) in ninety-five fathoms, instead of 40.3° which we found to be the general tempera- ture at that level (Stations 11, 41), though at one Station near by (12b) 94 bulletin: museum of compakative zoology. we got a bottom reading of 39.2°. In the deep basin off Cape Ann, Verrill's readings, in ninety, one hundred and eighteen and one hun- dred and fourteen fathoms, are 40°, 43°, 39° and 39°, at three stations near together. But the fact that we found a thick layer of bottom water very uniform in temperature in this region, suggests that the discrepancy in his readings was due to the faulty instruments. And it is at least suggestive that the average of his four readings in the basin is 40.2°, i. c, within .1° of our observations. Off Cape Cod, too, in 142 fathoms, close to Station 43, the bottom temperature in 1874 was 39° or 42°, agreeing fairly well with our record of 41.3° at Station 43. And the difference in depth is not significant in this case, because we encountered the uniform bottom water at 50 fathoms. On the other hand Verrill records a bottom temperature of 52° in 100 fathoms southwest of Jeffrey's Bank, where in 1912 the bottom reading, to judge from neighboring stations, must have been little, if any above 40.3°. And our entire experience makes it so improbable that the 100 fathom temperature is as high as 50° anywhere in the Gulf, that such a reading is best credited to the unreliability of the instrument with which it was taken. On the whole the bottom temperatures in Massachusetts Bay, in the western basin, and in the trough west of Jeffrey's Ledge were practically the same in 1873 as they were in 1912. But Verrill's readings for the northeast corner of the Gulf are so consistently lower than ours, that it is probable that the bottom water in that region actually was from 1° to 3° colder in 1873 and 1874 than it was in 1912. His records for 1874, (1875, p. 413) agree in a general way with our work in 1912, but as the same imreliable thermometers were used, and only one reading taken at each station, it is unwise to lay stress on them. Dickson's, (1901) charts show the salinity of the eastern half of the Gulf as below 32%o the Bay of Fundy 31%o or lower, and Massachu- setts Bay as below 32 for August, 1897 (no salinities are given for the remainder of the Gulf for these months). But on examining his tables, which give the tests of the water samples on which the charts are based, I did not find a single record from within the Gulf for either month, which suggests that the salinity credited in his charts to the eastern half of the Gulf was deduced from the low salinities revealed by several water samples taken in that month off the Nova Scotian Coast. But our own records show that his reconstruction of this region was probably incorrect, because it is certain that in August, 1912, there was an indraught of Atlantic water with salinities of 32.8 or more into the eastern part of the Gulf, and we have no actual data BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 95 to disprove the supposition that this is an annual, if not a constant phenomenon. A similar indraught is shown by Schott, (1902) on his chart of the Atlantic. Unfortunately salinities at other times of year do not aid as to whether or not the 1912 conditions were normal, for there are only two titration records from within the Gulf, north of George's Bank in Dickson's tables, one of 32.9 oif Cape Cod, April, 1896, the other of 32.3 off Cape Sable in the same month. There are several records in his table from George's Bank, and I have received two samples from its northern edge, collected November, 1911, with salinities of 32.7 and 32.9 respectively. General considerations. ' Various explanations have been proposed to account for the band of cold water of low salinity which bathes the coastal slope from New- foundland to Cape Hatteras, one of the earliest being that it is a branch of the Labrador Current flowing southerly along the shore. And although there is little actual evidence, other than low tempera- ture, in its support, this is the one w^hich has found its way most gener- ally into literature, scientific as well as popular. Thus Libbey (1891), in his discussions of ocean temperatures south of Nantucket, constantly refers to the cold wall as the "Labrador Current." Of late years, however, practical oceanographers have found less to recommend it, and Verrill, (1874) long ago questioned whether the low bottom tem- peratures which he observed off Portland in 1873 were not really a part of the cold bottom water of the North Atlantic rather than evidence of Arctic water. The facts, according to Verrill, do not warrant the assumption that an Arctic Current, properly so called, as dis- tinguished from tidal currents, enters the Gulf of Maine; but he quali- fies this generalization by adding that the Gulf gets constant acces- sions by the tides of cold w'ater which has primarily come from the north. According to Schott, (1897) and Hautreux, (1910) the source of the cold water, as far south as New York, is not the Labrador Current, but the St. Lawrence. But Pettersson, (1907) discarding the idea of an Arctic Current, definitely classes the cold wall along the North Amer- ican coast as "an updrift of the cold bottom water of the ocean when pushed against the coast banks," the motive force for this push being the "sinking cold water at Newfoundland," though, as he points out, " we know too little of the hydrography of the Gulf Stream and of 96 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. the cold wall on the American side of the Atlantic to be able to trace with security the origin of its waters." Quite a different explanation for the cold wall is proposed by Tizard, (1907, p. 343) who believes that the chief factor in forming the cold coast water is the discharge of fresh water from the rivers along the American coast, by which means large quantities of cold fresh water and fresh ice are emptied over the coastal slope. And he argues that neither upwelling of oceanic bottom water, nor the Labrador Current, has anything to do with the formation of the cold wall. The partial isolation of the Gulf of Maine from oceanic waters by the sill formed by George's and Brown's Banks, makes it possible that its cold waters need a different explanation from those of the "cold wall" west of Cape Cod; and the discussion of the latter is best post- poned until we have a better knowledge of their salinity. But so far as the Gulf is concerned, we can safelv sav that the low salinities in July and August certainly show that its waters are not predominantly Atlantic abyssal water welling up over the continental slope, because the salinity of the bottom water over most of the North Atlantic is about 34.9 (Murray and Hjort, 1912). The same index, salinity, shows that Tizard has suggested a factor of real importance, for besides the fresh water emptied into the Gulf of Maine annually' by its rivers (p. 90) there is also the annual rainfall of about 40 inches, a total annual increment of fresh water, which would make a layer more than a fathom thick over the entire Gulf. To offset this, there is the annual evaporation; and while this is not exactly known for any off-shore station in the Gulf, conditions on the neighboring coasts indicate that it is probably less than the rainfall. Rainfall and inflow from rivers combined are likewise considerably in excess of the annual evaporation all along the coast of Nova Scotia where the salinity, according to both Dickson, (1901) and Schott, (1902) is 32%o orless. The Gulf of St. La^vTence, has, of course, been mentioned by pre- vious authors as a source of fresh water, but its importance must be greater than has been usually recognized, because of the enormous ■extent of its watershed, including the St. Maurice, Saguenay, Humber, .and other large rivers, besides the St. Lawrence itself. Its rainfall, too, exceeds evaporation. The little that is known about the currents in its two mouths (Dawson, 1910) shows that its main outlet must be through Cabot Straits, as Schott represents it in his chart of ocean currents, (1902, pi. 39) not through the Straits of Belle Isle. The comparatively fresh St. Lawrence water is continuous with the water BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 97 with salinities of 32 or less, along the east coast of Nova Scotia. And if the Gulf of Maine receives any regular accessions of northern water of low salinity, it is probably from the Gulf of St. Lawrence, not from the Labrador Current. The temperatures of the Gulf of Maine are, of course, very low in comparison with the Gulf Stream off shore; and its surface tempera- ture, at least, is considerably lower than the average for its latitude, about 57°, as calculated by Kriimmel, (1904) as against a probable yearly mean of about 48° for the Gulf. But we must remember in this connection that on the east coast of North America cyclonic at- mospheric disturbances move as a whole from the land out over the sea, not from sea to land, as they do over Western Europe, and con- sequently, that the coastal waters may be expected to take their tem- peratures from the land climate instead of the latter being governed by oceanic temperatures, as is the case in Europe. If the Gulf of Maine were an enclosed basin, we would expect its bottom temperature to be about the same as the mean annual tempera- ture of the surrounding land-mass, just as Nordgaard, (1903) has found it for the Norwegian fjords. And as a matter of fact, the lowest temperatures which we encountered in the Gulf are practically the same as the mean annual for northern New England, i. c, that portion of the land mass from which the chilling winds of autumn and winter blow. The considerable snowfall must likewise be an active factor in chilling the surface water in winter, while the inrush of fresh snow- water, only a few degrees above freezing point, in spring, may be expected to show its effect in retarding the warming of the coast water as the season advances. Furthermore, the considerable thickness of the bottom water of uniform temperature in the western part of the Gulf, is good evidence of winter cooling, while our observations show that the temperature was lowest in the western half, just where cooling land winds and snow are most active, instead of in the eastern, where a northern current might be expected to show itself most clearly. Thus Verrill was probably correct in his contention that the waters of the Gulf are not abnormally cold, considering their geographic location, and the climate of the neighboring land mass. The possibility that cold northern water enters our Gulf in small amounts is not forbidden by the conclusion that the low temperature of the latter is chiefly due to winter cooling. On the contrary, the fact that the bottom temperatures on the coastal banks along the coast of Nova Scotia are much lower than at corresponding depths in the Gulf or further west, and that they decrease from southwest to north- 98 bulletin: museum of comparative zoology, east, as found by the Albatross in 1883 and 1885, together with the saUnities, as pointed out above (p. 97) is good evidence that there is a flow of St. Lawrence water along the coast of Nova Scotia toward the southwest. And, finally, at least two wreck courses (Hautreux, 1910) have been recorded with a southerly drift near Nova Scotia. But there are no wreck tracks nor iceberg tracks leading from the grand banks of Newfoundland toward Nova Scotia, such as might be expected were there any pronounced westerly drift of the Labrador current. The occasional occurrence of Arctic pelagic organisms in Massachusetts Bay and the Bay of Fundy, such as the medusa Ptychogena and the ctenophore Mertensia, neither of which has been able to establish itself in the Gulf, shows that there are occasional indraughts of the St. Lawrence water into the latter. But the fact that last summer the indrift was of Atlantic not St. Lawrence origin (p. 94), and the occasional record of tropical organisms, e. g., the siphonophore Physalia at Grand Manan, show that its influence is either sporadic, or seasonal, not constant. If any general conclusion can be drawn from the scanty oceano- graphic data yet available, it is that the Gulf of Maine owes its low temperature and salinity largely to local causes; i. e., to its geographic position and partial isolation by the sill formed by George's Bank; and that though there was an influx of ocean water in the summer of 1912 from the edge of the Gulf Stream, in other years, or at other seasons, there are more or less sporadic indraughts of cold water flow- ing from the northeast. This water, however, probably has no con- nection with the Labrador Current, but comes from the St. Lawrence. Preliminary notes on the plankton. The following notes on the macroplankton, preliminary to the spe- cial reports on the various groups, are offered because no attempt seems to have been made to study the pelagic fauna of the Gulf as a whole; and because the collections and oceanographic data of the Grampus allow a correlation between its plankton at a given time and the physical factors of the water, at the same time and place. With these ends in view, our main efforts were directed toward quali- tative, rather than quantitative results, though we devoted as much attention to the latter as was practicable. The usual program of plankton work during the day time, was to use the no. 20 (bolting silk) net at or near the surface, and to tow the coarse four-foot net hori- BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 99 zontal lor half an hour at some intermediate depth. When stations were occupied after dark, we usually used the four-foot net within a fathom or so of the surface, in this way getting very rich tows. The data of the hauls is listed in the table of stations (p. 135). The hauls with the quantitative (Hensen) net are discussed separately (p. 127). By far the most important member of the animal plankton over most of the Gulf, numerically at least, was the small copepod Calanus finmarckicm, which was taken at every Station (p. 115). This species plays much the same role in the ^^tal economy of the Gulf as it does in Norwegian waters on the other side of the Atlantic, being the chief food for pelagic fishes, particularly the mackerel. It is well known to fishermen under the name of "red feed," from the reddish color of a mass of these little crustaceans. At times it occurs in almost un- believeable numbers ; for example our four-foot net hauls in Massa- chusetts Bay near Cape Ann in July often yielded two or three quarts of this Calanus. At this time the plankton of the Bay was almost exclusively composed of C. finmarchicus, with very few other copepods; e. g. Pseudocalanus, Eurytemora, and Metridia; an insignificant num- ber of Sagittae (chiefly S. elegans); a few larval schizopods; an occa- sional full-grown schizopod {Mcganydiphanes norvcgica), and a few medusae, e. g., Aurelia, Cyanea, Melicertum, and the northern ctenophore, Bolinopsis infundibulum. We also obtained one specimen of the large pteropod Clione limacina in the Bay, and others off Cape Ann (p. 119). In the northeastern corner of the Bay, this general type of plankton was varied by the presence of great numbers of fish eggs (Station 1), and our several stations in the northeast corner of the Bay yielded many pelagic larvae of the cunner (Tautogolabrus), cod {Gadus coUarius), witch flounder {Glyptocephalus cytioglossus) , and sanddab {Hippoglossoides platessaides), with a few silver hake (Merluccius), redfish {Schasies marinus), haddock {Melanogrammus aeglifinus), rockling (Enchelyopus) and other species (p. 107). Twelve miles or so off Cape Ann (Station 2) there were very few fish eggs; and no fry; and over the western arm of the deep basin (Station 7) there were no eggs at all, but a considerable number of fish larvae, mostly cod, of which twenty-nine specimens were taken. The Calanus swarm, however, was nearly as dense as in the Bay; and we noted here, for the first time, the large boreal copepod Euchaeta norvegica, between seventy-five fathoms and the surface. There were no other copepod species in the haul. At this station we likewise cap- tured two large Meganyctiphancs norvegica, and one specimen of the 100 bulletin: museum of comparative zoology. pelagic boreal amphipod, Euthemisto, which was taken frequently from this point on, while a swarm of Sagitta elegans gave a new aspect to the tow. Clione, too, was represented by several large specimens. There were neither Aurelia nor Cyanea so far offshore; but the four- foot net yielded several large Beroe cucumis, a cosmopolitan form already often recorded from the Gulf. Perhaps associated with the abundance of Calanus, were the numerous Wilson's petrels which surrounded the ship as soon as we hove her to at this station. From Station 7 we ran in shore again, and worked for two days in Ipswich Bay, a region where I had previously found an abundant plankton, and which is proverbial for whales, sharks, etc., and the seat of an im- portant winter fishery. Calanus finmarchicus was still the prevalent organism, the nets bringing back a swarm of juveniles, besides several Euchaeta norvegica, great numbers of Sagitta elegmis (Stations 8, 9, 10, 11, 12b), Tomopteris helgolandica, represented by a very large specimen in the quantitative haul at Station 11, and, among Medusae, Aurelia and Cyanea in large numbers, with a few Mdicertum campan- ula, and Phialidium languidum. The latter species we found very widely distributed in the coastal waters of the Gulf. But the most important feature of Ipswich Bay, to us, was the immense number of pelagic fish eggs, largely Urophycis chus: and a haul of the eight-foot beam trawl for thirty minutes at Station 8 yielded the following large haul of fishes; twelve skates (Raja radiata) two Aspidophowides monoptcrygias, four Zoarccs anguillaris; twenty silver hake {Merluccius hilinearis,) two hake {Urophycis regius), thirty -four squirrel hake,- {Urophycis chus), two rocklings {En- chelyopsis cimhrius), forty-one sanddabs {Hippoglossoidcs plates- soides), six rusty flounders, {Limanda fcrrugnea), forty-eight witch flounders {Gly ptoccphalus cynoglossus), and seven large goosefish {Lo- phius piscatorius). The squirrel hake (C/rop/i?/cis chus) were full of ripe eggs and milt; and comparison of their eggs, fertilized on board, with the pelagic eggs taken in the tow, established the identity of the most abundant of the latter as belonging to this species. This dis- covery is of great interest, because very little is known of the early stages of any members of this genus, and nothing of this particular species. It, and the other fishes will be described by Mr. W. W. Welsh. Meantime it may be noted here that the fish were spawning in twenty- two fathoms, temperature 42.4°, salinity 32.39%o. In spite of the great numbers of pelagic eggs, Ipswich Bay and the waters immediately to the north yielded but few fry, except for the sanddab (Hippoglos- soidcs), of which twenty-four specimens of 10-22 mm. were taken at BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 101 Station 11. In Kittery Harbor, however, we obtained great numbers of Tautogolabrus and Merluccius. Our enforced return to Gloucester for repairs on July 18 gave us an opportunity to compare the plankton off the Harbor mouth (Sta- tion 12) with what we had found a week or two pre\'ious, with the result that there had been no appreciable change, the waters still being filled with the Calanus swarm besides an occasional Euchaeta, and a few fish eggs, and many fry, as noted above. Our run from Cape Ann to Casco Bay showed that the spawning area of the squirrel-hake, admitting our identification of the pelagic eggs to be correct, extended over the whole coast band, large hauls of fish eggs, including this species, being made at Stations 14 and 20. At the latter many cunner eggs (Tautogolabrus) were also taken; and a fevr eggs profebly belonging to the mackerel, which were school- ing in small numbers in this neighborhood at the time. ^lackerel eggs were likewise taken in our surface tows at our anchorage at Orr's Island, on August 1. Only two species of fish fry were taken in num- bers in the northwest corner of the Gulf and in Casco Bay. Most important of these, because of its purely boreal habitat, is the redfish (Sebastes marinus), no less than 320 larvae of which were taken in the closing net and in the intermediate haul at Station 19 (p. 108). At Station 22, likewise, it was represented by fifty-three specimens, in the open net haul. In Casco Bay, larval cunners (Tautogolabrus) were numerous. Along this stretch of coast we continued to find Calanus finmarchicus in large numbers, with a few Euchaeta norvegica; and at Station 13 we captured a few of the large blue copepod Anomaloccra patersoni on the surface, a species frequently taken after this, occasionally in large numbers (p. 118); and several other copepods in lesser numbers, as shown in the table (p. 115). Off Cape Porpoise we first encountered the amphipod Euthemisto in large numbers. Here, too, our tows revealed many specimens of the pteropod Limacina balea; while Stations 19 and 22 added a fresh Chaetognath, Sagitta serratodentaia in small numbers. Another addition to the plankton, in this region, was the large hydromedusa Staurophora mertensii, which we first met at Station 14, where three large specimens were taken in a haul of the four foot net from twenty fathoms. Meganydiphanes norxegica, too, occasionally occurred in our hauls off Casco Bay, (Station 19). In the coast region Aurelia and Cyanea were taken in most of the hauls, but usually not on the surface; though several large specimens were seen floating at Station 22. Our most notable find in this region was 102 bulletin: museum of comparative zoology. four fragmentary specimens of the hydromedusa Halopsis ocellata, taken at Stations 15, 22, and 23. This species, first discovered in Massachusetts Bay (A. Agassiz, 1865, p. 102) has since been recorded only once, by Fewkes, (1888), who found it in considerable numbers "near the wharves at Grand Manan." The chief point of interest about this species is its otocysts, for though Agassiz figured them (1865) it has remained questionable whether they are 5pen or closed, and consequently Browne, (1910) found it impossible to refer the genus definitely either to the Mitrocomidae or to the Eucopidae. Fortu- nately our specimens, though much battered, show these organs well, and it is easy to demonstrate that they are open pits. The opening is evident on surface views of the oral side of the velum, and large enough to admit a fine bristle. Consequently Halopsis is a mitro- comid. The specimens agree with Fewkes's statement as to the inde- pendent origin of the radial canals from the stomach (in the original account they are described as arising in four groups) . Our run out to Piatt's Bank showed that very few fish were spawning except close to the shore, for the tows at Station 23, on the bank, contained no eggs at all, nor did we meet any over the deep trench a few miles further south (Station 24), while very few were found over Jeffrey's Bank (Station 25) except for a Lophius veil, with the eggs nearly ready to hatch, which we picked up from the surface at this station. And to complete the brief survey of fish eggs I may add that very few were taken at any of our stations further north or east; none at all at the off-shore stations over the Eastern Basin (Stations 27, 28) on German Bank (Stations 29, 30), off Lurcher Shoal (Station 31), off Mt. Desert Rock (Station 32) or in the Grand Manan Channel. On the other hand we captured 190 larval red fish (Sebastes) on Piatt's Bank (Station 23); 18 at Station 27, 61 at Station 28; and 27 at Station 32; but it was not taken on Jeffrey's Bank (Station 25); nor along the coast from Grand Manan to Penobscot Bay (Stations 33 to 39). At our off-shore hauls the plankton repeated, in a general way, the conditions met nearer land, Calanus with a few other copepods, notably Euchneta norvegica and Anomalocera patersoni, still forming the bulk of the hauls (Stations 23, 24, 27, 28). But the haul from twenty fathoms at Station 23 yielded an important addition to the list of copepods, in the Arctic Calanus, C. hyperboreus, represented by six specimens among the thousands of C. finmarchicus. We now met Meganyctiphanes more regularly, considerable numbers of this schizopod being taken at Station 27, 80-0 fathoms. And at Station BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 103 23 we found a single specimen of the medusa Tiaropsis diademata, which is abundant in Massachusetts Bay in June. At Station 23 we first met Pleurobrachia pileus, and we frequently took it later, further north and east; we saw Beroe cucumis on the surface, and captured sticklebacks, and a large isopod (Idotea) from floating Fucus. In these off-shore waters Sagitta serratodentata was more plentiful than we had found it before, — a case treated at length elsewhere (p. 121) and an occasional fragmentary agalmid was likewise taken (p. 121) besides considerable numbers of fish fry. At Station 27 Calanus finmarchicus was taken in swarms at the surface, the only time we found it abundant at that level, in the day time, though it often was at night. Euthemisto was plentiful at the off-shore Stations 29 and 31, and at the former we took one Tiara pileata, and two Aglantha (40-0 fathoms), this being the first time the latter was encountered during the cruise. On the other hand, we found none of the typical shore forms, e. g., Aurelia, Cyanea, Meli- certum; and over the Eastern Basin not even Staurophora, Phialidium, Beroe, Bolinopsis, or Pleurobrachia, though the last three, of course, are not dependent upon shallow water at any stage in development. German Bank proved interesting, for though the surface tempera- ture was low (52°, Station 30) and the bulk of the tow consisted of Calanus finmarchicus, with a few Euchaeta, Anomalocera, a large number of the schizopod Euphausia, the amphipods Hyperia galba and Euthemisto, Tomopteris helgolandica, Sagitta elegans, and S. serratodentata, forming a typical boreal assemblage, the surface haul also yielded two large Salpa fusiformis and two specimens of the siphonophore Physophora hydrostatica. During the next day Salpae were occasionally seen on the surface; and at Station 31 several were taken in the tow, all S. fusiformis (p. 121). But here, as on German Bank, the plankton as a whole was the same as we had found over the Gulf as a whole, Calanus finmarchicus composing far the chief bulk of the haul. This proved to be an interesting station, because the open net from fifty-five fathoms brought back several specimens of the cold water Chaetognath Eukrohnia hamata, a species found on the surface in Arctic and Antarctic regions, but limited to the mesoplankton in temperate and tropical latitudes. This same haul also yielded two specimens of the large Sagitta lyra; and neither of these species was taken again during the voyage. The list of copepods also received an addition, Euckirella rostrata. After leaving this station we saw no more Salpae. Twelve miles off Mt. Desert Rock, August 16, 3 A. M., we made a 104 bulletin: museum of compailvtiye zoology. rich surface haul of Calanus finmarchicus, with a few other copepods, e. g. Centropages, Metridia, Anamalocera, and Euchaeta, besides Meganyctiphanes nonrgica, Hyppolyte, Euthemisto, Limacina halea, Sagitta elegans and S. serratodentata, Tomopteris helgolandica, Clione limacina, Pleurobrachia, Phiahdium, and agalmid fragments, i. e., the plankton was of the same type as off shore and further west ; and rich quantitatively. But when we approached shore, off Moose Peak, our hauls were extremely barren, by far the poorest yet made. The four-foot net, hauled for three Cjuarters of an hour, at Station 33, with an electric light in its mouth, contained only a few Calanus, four medium sized Staurophora, and a few Sagittae, the whole, aside from the large Medusae, being less than 20 cc. in bulk. Tliis was quite the contrary to what we expected, as the northeastern corner of the Gulf and the Bay of Fundy have always been credited with a rich pela- gic life. But in the Grand Manan Channel (Station 34), the plankton was even poorer than at Station 33, the four foot net, hauled from SQ-O fathoms, containing almost nothing except a ^ery few Calanus and other small copepods, while a few Staurophora were seen on the surface. And much the same condition was encountered in the mouth of the St. Croix River, where surface tows were made on August IS, very little being taken, or seen, except Staurophora. In Eastport Harbor, however, many ^^leganyctiphanes, probably attracted by refuse from the sardine factories, were taken on the surface. "When we returned through Grand Manan Channel, we made a haul off the north end of Campobello Island, where the four-foot net did not bring back even a single copepod; but it yielded large numbers of Balanus eggs in segmentation stages; and a few Staurophora were seen on the surface. Xear the entrance of the Channel (Station 35) the water was hardly more productive, the whole catch of the four-foot net (35-0 fathoms), chiefly Calanus and Sagittae, being contained in an ordinary table spoon; while no Medusae or ctenophores were seen on the surface. That night, however, in Cutler Harbor, we found a fairly rich neritic plankton, chiefly copepods, gammarid amphipods, and the hydroraedusid Sarsia. When we once more ran off shore to the edge of the deep basin, August 20 (Station 36), the water was occupied by the Calanus swarm, with a few Euthemisto, a few Euchaeta, many Sagitta, chiefly S. elegans, Aglantha digitale, Beroe cucuinis, Mega- nyctiphanes, and Staurophora, i. e., a typical Gulf of Maine plankton in considerable quantity. And the richness of this station and that of Station 32, showed that the edge of the dense Calanus swarm followed the 100 fathom curve, the barren zone being only a narrow coast belt. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 105 At Station 38 and 39, the nets yielded comparatively little except diatoms (p. 133), though more than in Grand Manan Channel. In fact it was not until Penobscot Bay was passed that we once more ran into copepods in abundance near the coast. The poverty of the macroplankton in general was shared by the fish fry, for our nets did not yield a single young fish along this whole stretch of coast, i. e. Stations 33 to 39. At Station 40 we once more met a rich copepod plankton, chiefly Calanus finmarchicus on the surface as well as in the intermediate haul. Calanus hypcrhoreus was likewise represented by one specimen (20-0 fathoms). Considerable numbers of larval Se- bastes were taken at this Station; and swarms of Plciirohrachia pilous and Phialidium languidum gave the tow a distinctive character dif- ferent from any previously taken. Between Station 40 and Cape Ann (Station 41), the Calanus swarm was once more met, but at this Station there were about as many Centropages as Calanus on the surface; and a surface haul at night off the Cape (Station 42) yielded large masses of Calanus. The tow at this Station was notable for containing large numbers of the copepod Auomalocera patersoni, be- sides Euthemisto, Tomopteris hclgolandica, Sagitta clcgans, Cyanea, Staurophora, Phialidium, and many fish larvae. The plankton oft' Cape Cod at the end of August (Station 43) proved to be of the same type that we had found generally over the Gulf, the prevailing ani- mal being Calanus fijimarchicm, with Eucheata norvegica in less abun- dance; Euthemisto, Pleurobrachia, Beroe, Staurophora, and a few larval fishes were also taken. Our lines do not afford any informa- tion as to how far south the Calanus swarm extended; but some tows made by Capt. John McFarland of the fishing schooner Victor revealed this copepod in great numbers five miles east of Chatham, on September 20. However, twelve miles S. E. of Chatham, a day or two later, his tow shows that it was outnumbered by Pseudocala- nus, five hundred to one. And, as pointed out (p. 121) he collected a pure Salpa plankton on the surface twenty-five miles oft' the same port on September 30, which is good evidence that Gulf Stream water was making its influence felt in that region. Off Cape Ann (Station 42) fish fry of several species, notably cunner (Tautogolabrus), redfish (Sebastes), rockling (Enchelyopus) and witch flounder (Glyptocephalus) were taken; and in the southern half of Massachusetts Bay (Station 44) the hauls yielded many larval sanddabs (Hippoglossoides) and witch flounders (Glyptocephalus), with a few redfish (Sebastes), silver hake (Merluccius), and rockling (Enchelyopus). The hauls off Cape Cod (Station 43) contained only nine fish fry, five Sebastes, and four Enchelyopus. 106 bulletin: museum of comparative zoology. Of the three components, Arctic, Boreal, and Temperate Atlantic, into which the noi'thern pelagic communities can be divided according to Hjort (Murray and Hjort, 1912, p. 637), the plankton of the Gulf belongs distinctively to the Boreal, for only a single species distinc- tively characteristic of polar waters, Calanus hyperborcus, was detected in 1912. Thus the ctenophore Mcrtensia ovum, was conspicuously absent, though it is known from Massachusetts Bay (A. Agassiz, 1865) and is recorded from the Bay of Fundy by Fewkes, (1888). The polar pteropod Limacina helicina was likewise wanting, whereas its boreal relative L. balea was taken at several stations, in some abundance. Nor did we detect the Arctic prawn, Hymenodora glacialis, a species lacking in boreal as well as in tropical waters. On the other hand Calanus finmarchicus, the most characteristic animal of all in the Gulf, is the most important member of the Boreal, as opposed to the polar plankton, in the Norwegian Sea and in the North Sea; and it is the commonest copepod off San Diego, California (Esterly, 1905, p. 126). Euthemisto, Meganyctiphanes nortegica, and Euchaeta norvegica are all characteristic of the Norwegian Sea, and of the southern edge of the Newfoundland Banks (Murray and Hjort, 1912, p. 108). Clione limacina, too, is by no means a sure indication of polar water, for though it is abundant in the Labrador Current off the east coast of Newfoundland, and has been taken off the west coast of Greenland, near Spitzbergen, and at other, Arctic stations, it is not associated with polar water in the Norwegian Sea, (Murray and Hjort, 1912, p. 107) but, on the contrary, is found in Atlantic water there, and south of Iceland. To judge, however, from its great abundance in high latitudes and comparative scarcity in our Gulf, it appears to reach its maximum development in a lower temperature than that of the Gulf of Maine in summer. And neither is Eiikrohnia hamata purely Arctic, for it occurs in the mesoplankton at lower latitudes; as for example in the Bay of Biscay, where Fowler, (1905) found it in one haul from fifty fathoms, i. e., at about the same depth as our one record, and in many hauls from greater depths. And there is no more reason to assume a polar origin for the Gulf of Maine speci- mens than there is for the Biscayan ones. Most of the important Medusae and ctenophores, for example Aurelia, Cyanea, Melicertum, Bolinopsis septentrionalis, are regular inhabitants of the Norwegian Sea, and of the northern part of the North Sea. Staurophora is known from Helgoland; while Pleuro- brachia pileus and Beroe cucumis are apparently cosmopolitan. To- moptcris hclgolandica is known from the North Sea, the coast of BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 107 Norway, the English Channel, the northeast coast of Scotland, and from the Grand Banks of Newfoundland; and Sagitta elegans is a characteristic member of the North Sea plankton. Most oceanic species so far detected in the collections, e. g., Salpa mucronata, and 8. fimformis, Sagitta scrratodcntata, Agalma elegans, Physophora Iii/drostatica, are dwellers in warm or in temperate waters, the only far northern records of any of them being obviously the result of warm currents (for northern records of Salpa, see Apstein, 1909: of Sagitta scrratodentata, see Ritter Zahony, 1911; Agalma and Physophora, see Bigelow, 191 1 . And the resemblance which the Gulf bears in a small way, to the Norwegian Sea in the more important constituents of its zoiiplankton, is heightened by the fact that Salpa, Agalma, and Physophora are regular summer visitors to the latter with the northward movement of Atlantic water (Helland Hansen and Nansen, 1909, Murray and Hjort, 1912), while their presence in our Gulf is positive evidence of an influx of water from the northern edge of the Gulf Stream. List of fishes. (Identified by W. W. Welsh, U. S. Bureau of Fisheries). 1. Larval and postlarval stages taken in the plankton hauls. Argentinidae. Smelt. Osmerus mordax (Mitchill). Portland Harbor July 31 Surface 1 specimen 19.5 mm. Herring Smelt. Argentina silus Ascanius. Station 27 August 14 35 fathoms 1 specimen 49 mm. Gasterosteidae. Three-spined Stickleback. Gasterosteus aculeatus Linne. Station 11 30-0 fathoms 1 specimen 2.3 cm. Station 23 surface 4 specimens 3.9-3 cm. Station 25 sm-face 8 specimens 3.9-2.8 cm. August 13 surface 4 specimens 4.4-3.7 cm. Station 29 surface 1 specimen 4.6 cm. Station 30 surface 11 specimens 4.2-2.8 cm. August 16 surface 1 specimen 4.2 cm. Station 43 surface 2 specimens 3.3-2.8 cm. 108 bulletin: museum of comparative zoology. Two-spined Stickleback. Gasterosteus hispinosus Walbaum. Station 23 surface 9 specimens 3-2 . 6 cm. August 13 surface 10 specimens 3 . 4-2 . 7 cm. Station 29 surface 1 specimen 4.5 cm. Syngnathidae. Pipefish. Siphostoma fuscum (Storer). Portland Harbor July 31 surface 1 specimen 14.5 mm. Cunner. Station 5 Gloucester Harbor Station 12 Kittery Harbor Orr's Island, Me. Casco Bay Station 42 Station 44 Labridae. Tautogolabrus adspersus July 12 surface July 19 surface July 22 5-0 fathoms July 27 surface 12 19 22 27 July 30 August 4 August 24 August 31 surface surface 20-0 fathoms 20-0 fathoms (Walbaum) (?). 25 specimens 7 specimens 6 specimens 102 specimens 35 specimens 80 specimens 3 specimens 1 specimen 4.5-2.5 mm. 6.5-5 mm. 6.5-5 mm. 6.5-3.5 mm. 6.5-3 mm. 5-2 mm. 9-6.75 mm. 8.5 mm. SCORPAENIDAE. Redfish. Sebastes marinus (Linne). Station 12 Station 14 Station 19 Station 19 Station 22 Station 23 Station 27 Station 27 Station 28 Station 31 Station 31 Station 32 Station 40 Station 42 Station 43 Station 44 July 22 July 24 July 29 July 29 August 7 August 7 August 14 August 14 August 14 August 15 August 15 August 16 August 22 August 24 August 29 August 31 5-0 fathoms 20-0 fathoms 20 fathoms 25-0 fathoms 30-0 fathoms 20-0 fathoms 35 fathoms 80-0 fathoms 30-0 fathoms 55-0 fathoms 25-0 fathoms surface 20-0 fathoms 20-0 fathoms 35-0 fathoms 25-0 fathoms 1 specimen 6 specimens 150 specimens 170 specimens 53 specimens 190 specimens 5 specimens 13 specimens 61 specimens 2 specimens 13 specimens 27 specimens 20 specimens 5 specimens 5 specimens 6 specimens 12 mm. 11-6.5 mm. 9-6.5 mm. 9.5-6.5 mm. 8.5-6.5 mm. 13.5-7 mm. 20.5-13 mm. 21-13 mm. 16-7.5 mm. 8-7 mm. 12.5-9 mm. 15-7 mm. 13-7.5 mm. 12-8.5 mm. 12.5-9 mm. 11.5-7 mm. Station 19 Station 21 COTTIDAE. Artediellus atlanticus Jordan and Evermann. 40-0 fathoms 2 specimens 6.3-4.3 cm. 60-0 fathoms 4 specimens 5.1-4.2 cm. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 109 Cyclopteridae. Lumpfisli. Cyclopterus lumpus Linne. Station 25 August 8 surface 16 specimens 91-13 mm. Station 26b August 13 surface 53 specimens 57-10.5 mm Station 27 August 14 surface 1 specimen 44 mm. Station 30 August 14 surface 9 specimens 70-21 mm. Between Petit Ma- nan and Libbey Island August 16 surface 39 specimens 34-14 mm. Station 40 August 22 surface 1 specimen 10.5 mm. Liparididae. Liparis liparis (Linne) (?). Station 12 July 22 5-0 fathoms 9 specimens 9-5.5 mm. Station 44 August 31 25-0 fathoms 2 specimens 7-5.5 mm. Blenniidae. Pholis gunnellus (Linne). Station 25 August 8 surface 1 specimen 39 mm. August 13 surface 1 specimen 29 mm. Uharia subbifurcata (Storer) (?). Station 5 July 12 surface 2 specimens 8 mm. Station 12 July 22 5-0 fathoms 8 specimens 14-8 mm. Station 14 July 24 20-0 fathoms 3 specimens 12-10 mm Station 20 July 31 7-0 fathoms 1 specimen 14 mm. Station 42 August 24 20-0 fathoms 1 specimen 15.5 mm. ZOARCIDAE. Lycenchelys verrillii (Goode and Bean) . Station 21 60-0 fathoms 1 example 10 cm. Merlucciidae. Silver Hake, Merluccius bilinearis (Mitchill) (?). Station 5 Kittery Harbor, Me. Orr's Island, Me. Station 44 July 12 July 27 July 30 August 31 surface surface surface 20-0 fathoms 8 specimens 22 specimens 2 specimens 9 specimens 4-2.5 mm. 4-2.5 mm. 3 mm. 10-6 mm. 110 bulletin: museum of comparative zoology. Gadidae. Cod. Gadus callarias Linne. Station 7 July 16 18-0 fathoms 29 specimens 15-4.5 mm, Station 11 July 17 25-0 fathoms 1 specimen 8.5 mm. Station 12 July 22 5-0 fathoms 61 specimens 19.5-8 mm, Haddock. Melanogrammus aeglifinus (Linne) (?). Station 12 July 22 5-0 fathoms 6 specimens 21-10 mm. Station 16 Station 25 Station 27 Station 30 Station 31 Station 5 Station 11 Orr's Island Casco Bay Station 25 Station 29 Station 30 Station 41 Station 42 Station 43 Station 11 Station 12 Station 20 Station 42 Station 44 Station 12 Hake. Urophycis sp. July 26 August 8 August 14 August 14 August 15 surface surface surface surface sm^ace 1 specimen 1 specimen 1 specimen 1 specimen 1 specimen 84 mm. 70 mm. 67 mm. 41 mm. 102 mm. Reckling. Enchelyopfus dmbrius (Linne). July 12 July 17 July 30 August 4 August 8 August 14 August 14 August 24 August 24 August 29 surface 30-0 fathoms surface surface surface surface surface sm-face surface 35-0 fathoms 5 specimens 6 specimens 1 specimen 1 specimen 1 specimen 1 specimen 1 specimen 1 specimen 2 specimens 4 specimens Enchelyopus cirnbrius (Linne). (?). July 17 July 22 July 31 August 24 August 31 25-0 fathoms 5-0 fathoms 7-0 fathoms surface 25-0 fathoms 4 specimens 2 specimens 1 specimen 6 specimens 2 specimens 5-3 mm. 38-12 mm. 2 mm. 5.5 mm. 42 mm. 20 mm. 31.5 mm. 23 mm. 16.5-13.5 mm. 44-39 mm. 11-5 mm. 5.5 nma. 5 mm. 10-5 mm. 8.5-5 mm. Cusk. Brosme hrosme (Miiller) (?). July 22 5-0 fathoms 1 specimen 13.8 mm. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. Ill Station 11 Gloucester Harbor Station 21 Pleuronectidae. Sanddab. Hippoglossoides platessoides (Fabricius). July 17 25-0 fathoms 24 specimens 22.5-10 mm. July 18 August 2 surface 60 fathoms 1 specimen 1 specimen 23.5 mm. 89 mm. Station 44 Hippoglossoides platessoides (Fabricius) (?). August 31 25-0 fathoms 24 specimens 9.5-6 mm. Winter Flounder. Pseudopleuronedes americanus (Walbaum) (?). Station 7 July 16 18 fathoms 1 specimen 10.5 mm. Station 11 July 17 25-0 fathoms 1 specimen 13 mm. Station 20 July 31 7-0 fathoms 1 specimen 6.5 mm. Witch Flounder. Glyptoccphalus cynoglossus (Linne). Station Station Station Station Station Station No label 12 July 14 21 30 42 44 Station 7 Station 12 22 July 24 August 2 August 14 August 24 August 31 5-0 fathoms 20 fathoms 60 fathoms surface 20-0 fathoms 25-0 fathoms 2 specimens 1 specimen 2 specimens 8 specimens 1 specimen 6 specimens 20 specimens Pleuronedids unplaced. July 16 18 fathoms 1 specimen July 22 5-0 fathoms 15 specimens 16.5-8.5 mm. 9.5 mm. 16.5-8 mm. 108 -65 mm. 18.5 mm. 18-10 mm. 16.5-9 mm. 7 mm. 10-6 mm. Station 5 LOPHIIDAE. Goosefish. Lophius piscatorius Linne. July 12 Surface 1 specimen 6.5 mm. Station 16 Station 6 Station 8 Adult stages taken in the trawl. Rajidae. Little Skate. Raja erinacea Mitchill. 25 fathoms 1 specimen Prickly Skate. Raja radiata Donovan. 27 fathoms 22 fathoms 1 specimen 12 specimens 112 bulletin: museum of comparative zoology. Station 1 Station 6 Station 13 Station 15 Station 16 Station 19 Station 21 Station 23 Station 1 Station 6 Scorpaenidae. Redfish. Sebastes marinus (Linne). 33 fathoms 27 fathoms 30 fathoms 30 fathoms 25 fathoms 50 fathoms 60 fathoms 47 fathoms 1 specimen 14 specimens 1 specimen 2 specimens 6 specimens 1 specimen 7 specimens 5 specimens COTTIDAE. Triglops 2>ingelii Reinhardt. 33 fathoms 27 fathoms 1 specimen 1 specimen Sculpin. Myoxocephalus octodecimspinosus (Mitchill), Station 1 Station 6 Station 17 Station 19 33 fathoms 27 fathoms 16 fathoms 50 fathoms 1 specimen 5 specimens 1 specimen 2 specimens Sea Sculpin. Hemitripterus americanus (Gmelin). Station 15 Station 21 30 fathoms 60 fathoms 1 specimen 1 specimen Agonidae. Alligator Fish. Aspidophoroides monopterygius (Bloch). Station 1 Station 6 Station 8 Station 15 33 fathoms 27 fathoms 22 fathoms 30 fathoms 8 specimens 9 specimens 2 specimens 1 specimen Station 6 Station 16 Blenniidae . Ulvaria subbifurcata (Storer). 27 fathoms 25 fathoms 1 specimen 1 specimen BIGELOW: EXPLOKATIONS IN THE GULF OF MAINE. 113 Station 6 Station 8 Station 15 ZOARCIDAE. Eelpout. Zoarces anguillari^ (Peck). 27 fathoms 3 specimens 22 fathoms 4 specimens 30 fathoms 7 specimens Station 1 Station 8 Station 15 Station 16 Station 21 Merlucciidae. Silver Hake. Merluccius bilinearis (Mitchill). 33 fathoms 22 fathoms 30 fathoms 25 fathoms 60 fathoms 1 specimen 20 specimens 2 specimens 1 specimen 9 specimens Station 1 Station 6 Station 15 Station 16 Station 8 Station 17 Station 21 Station 8 Station 13 Station 15 Station 16 Station 21 Station 1 Station 8 Station 15 Station 16 Gadidae. Cod. Gadus callarius Linne. 33 fathoms 27 fathoms 1 specimen 2 specimens Haddock. Melanogrammus aeglifinus (Linne). 30 fathoms 25 fathoms 1 specimen 1 specimen Spotted Hake. Urophycis regius (Walbaum). 22 fathoms 11 fathoms 60 fathoms 2 specimens 3 specimens 1 specimen Squirrel Hake. Urophycis chus (Walbaum). 22 fathoms 30 fathoms 30 fathoms 25 fathoms 60 fathoms 34 specimens 2 specimens 4 specimens 1 specimen 2 specimens Rockling. Enchelyopus cimhrius JX^mne) . 33 fathoms 1 specimen 22 fathoms 2 specimens 30 fathoms 1 specimen 25 fathoms 2 specimens 114 bulletin: museum of comparative zoology. Pleuronectidae. Sanddab. Hippoglossoides platessoides (Fabricius). Station 6 Station 8 Station 13 Station 15 Station 16 Station 21 27 fathoms 22 fathoms 30 fathoms 30 fathoms 25 fathoms 60 fathoms 7 specimens 41 specimens 2 specimens 3 specimens 2 specimens 3 specimens Rusty Flounder. Ldmanda ferruginea (Storer). Station 6 Station 8 Station 15 27 fathoms 22 fathoms 30 fathoms 3 specimens 6 specimens 1 specimen Winter Flounder. Pseudopleuronectes americanus (Walbaum) Station 17 11 fathoms 6 specimens Witch Flounder. Glyptocephalus cy Station 1 33 fathoms 1 specimen Station 6 27 fathoms 1 specimen Station 8 22 fathoms 48 rpecimens Station 16 25 fathoms 1 specimen Station 21 60 fathoms 1 specimen Station 8 Station 15 Station 21 LOPHIIDAE. Goosefish. Lophius piscatorius Linn^. 22 fathoms 30 fathoms 60 fathoms 7 specimens 2 specimens 3 specimens BIGELOW: EXPLOR-'VTIONS IN THE GULF OF MAINE. 115 List of copepods. (Identified by C. 0. Esterly). Stations -^ Dekius — >• 2 60-0 4 15-0 6 0 6 27-0 7 75-0 11 25-0 12 40-0 14 20-0 15 15-0 16 15-0 19 25 19 25-0 20 7-0 Calanus finmarchicus Calanus hyperbortus Pseudocalanus elongatus Paracalanus parvus Euchirella rostrata Euchaeta norvegica Centropages hamatus Centropages typicus Temora longicornis Eurytemora herdmani Metridia lucens Anomalocera palersoni Acartia clausi Acartia longiremis Tortanus discaudatus Corynura discaudatus X X X X 1000 1000 X X X 50 100 50 40 X 1 1 3 3 3 1 X 1 1 1 1 1 1 X X 100 X ?. 1 Stations — > Depths — > 21 10-0 22 30-0 23 20-0 24 0 25 30-0 25 20-0 26 0 26b 0 27 0 27 80-0 27 35-0 28 25-0 29 20-0 Calanus finmarchicus Calanus hyperboreus Pseudocalanus elongatus Paracalanus parvus Euchirella rostrata Euchaeta norvegica Centropages hamatus Centropages typicus Temora longicornis Eurytemora herdmani X X 1000 6 X 100 20 X X X 500 500 X X 1 4 2 10 X X 5 1 2 Anomalocera palersoni X X 1 Acartia longiremis Corynura discaudatus 116 bulletin: museum or comparative zoology. Stations — > Depths — > 30 0 31 55-0 32 0 35 40-0 36 75-0 37 15-0 38 40-0 39 75-0 40 0 40 25-0 41 0 41 20-0 42 0 Calanus finmarchicus Calanus hyperboreus Pseudocalanus elongatus Paracalanus parvus X 50 X X X X 100 X X 50 1 1 40 50 100 1 .... 1 X 2 Euchaeta norvegica X X Centropages hamatus Centropages typicus TevfiOTa longicornis X 1 30 1 60 EuvyieTnova herdniani Metridia lucens X X 4 X X 1 Anomalocera patersoni Acctvtia clausi X X X 12 ActXTtia longiremis Tortanus discaudatus Corynura discaudatus Stations —>■ Depths —^ 42 15-0 43 35-0 43 85-60 44 0 44 20-0 U CQ 9 o 5 •a a o A. 8 B. 0 C. 10 Calanus finmarchicus Calanus hyperboreus Pseudocalanus elongatus Paracalanus parvus Vlurhirpllfi rostrata X 2000 1 1 50 X 20 X X 50 300 5 X X 20 X 1 10 500 Euchaeta norvegica Centropages hamatus Centropages typicus Temora longicornis Eurytemora herdmani 1 1 1 3 1 1 1 10 X 1 X 1 X 1 1 Anomalocera patersoni X X Acartia longiremis 1 X X Corynura discaudatus The numbers indicate the proportional abundance of species in a haul, not numerical occurrence. X indicates that the species oc- curred : — BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 117 Station A 6 miles off Cape Porpoise August 18 8 fathoms. Capt. John INIcFarland. Station B 8 miles E. of Chatham September surface. Capt. John McFarland. Station C 12 miles S. E. of Chatham September 10 fathoms. Capt. John McFarland. Distribution of the more important plankton species. Among the objects of the exploration of the Gulf is the correlation of the distribution, seasonal and geographic, of the more important members of the plankton with the physical characters of the waters in which they live; and the determination of the factors which govern their times of reproduction, movement, and abundance. Obviously the summer work in 1912 is only the first attack on the problem; but the data acquired is valuable because salinity and temperature are known for the various captures, and can be used as the basin of future work. In the following notes, the occurrence of some of the more important animals is summarized, without any reference to earlier records for the region. Calanus finmarchicus.- — As pointed out, (p. 99) this copepod was taken at every station, including the harbors of Gloucester, Kittery, and Portland; and it greatly predominated over all others at most of the off-shore stations. The exceptions, as noted above (p. 105), and in the table (p. 115), were the surface hauls at Stations 41 and 44, which yielded nearly equal numbers of Calanus and of Centropages; the closing net haul at Station 43, in which there were about as many Euchaeta as Calanus, and Capt. McFarland's haul twelve miles S. E. of Chatham, late in September, in which Pseudocalanus outnumbered Calanus one hundred to one. In twelve hauls the copepod constituent of the plankton was ex- clusively Calanus, e. g. in the northeastern part of Massachusetts Bay and off Cape Ann in July; and at the off-shore stations as a whole very few other copepods were found. Thus at Station 7, there were about 1,000 Calanus to one Euchaeta; at Station 23, about 1,000 Calanvs finmarchicus to six C. hypcrhoreus to four Euchaeta; at Sta- tion 27, 500 Calanus to two Euchaeta to two Metridia; at Station 28, pure Calanus; and at Station 43, in the open net, 2,000 Calanus to one Euchaeta; cf. table (p. 116). Calanus hyperboreus. This Arctic species was taken twice, at Sta- 118 bulletin: museum of comparative zoology. tion 23, 20-0 fathoms, six specimens, and Station 40, 20-0 fathoms, one specimen. The importance of these captures has been noted (p. 106). Euchacta norvegica. — This species has been detected at Stations 7, 8, 12, 23, 27, 31, 32, 36, 43, and from two miles off the Isles of Shoals, i. e., Massachusetts Bay, Ipswich Bay, both arms of the deep basin, off Cape Cod, on Platts' Bank, off the Nova Scotian Coast and the coast of Maine. Thus Euchaeta norvegica was generally distributed over the Gulf, though of irregular occurrence. The only localities where it was abundant were at Station 36, in a haul with the open net at seventy-five fathoms, and at Station 43, where many were taken in the closing net at eighty-five to sixty fathoms, forming the bulk of the haul. As most of the hauls were taken from intermediate depths, neither the horizon from which the specimens came, nor, consequently, the precise salinity and temperature can be determined. The known salinities and temperatures are: — Station 32 surface temperature 57° salinity 32.51%o. Station 43 85-60 fathoms temperature 42° salinity about 33.5%o. Euchaeta was only once taken in a surface haul (Station 32), and the fact that it was most abundant in deep hauls agrees with its occur- rence in the Norwegian Sea. Anomalocera patersoni. — This copepod was taken at Stations 13, 24, 26b, 27, 30, 37, 40, 41, 42, being thus generally distributed over the western side of the Gulf, in Frenchman's Bay, and on German Bank. But it was not taken in Massachusetts Bay, nor over the off-shore por- tion of the Gulf as a whole, its only occurrence far from land being at Station 24. It was taken in surface hauls, never in the closing net, only once (Station 37) in the open net from intermediate depths. The temperatures at which it occurred ranged from 52°-61°, the salinity from 31%o to 32.7%o- The fact that it was a purely surface form makes it probable that it was more widely distributed than our records show, for comparatively few hauls were made at the surface with the large net. But it was conspicuously absent from the surface hauls made at Stations 27, 28, 29, 31, 32, a fact showing that it is not brought to the Gulf by the indraught of oceanic water which is no- ticeable over the region covered by these stations. Meganydiphanes norvegica. — The following notes are based only on the occurrence of large adults of unmistakable identity; probably the list of localities will be largely augmented by identification of the large series of young schizopods. Thus restricted, Meganycti- BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 119 phanes was taken at Stations 3, 7, 19, 25, 26a, 27, 32, in Eastport Harbor, and two miles southeast of Duck Island, Mt. Desert: i. e., Massachusetts Bay, the coast of Maine, Jeffrey's Bank, the region off Casco Bay, and both sides of the deep basin. The oeeanographic data of the captures is as follows : — Station 25 surface temperature 56° salinity 32.34%o Eastport Harbor surface temperature 50° salinity about 32.5%o Station 27 closing net 40 fathoms temperature 45° salinity about 33.6%o Station 32 surface temperature 57° salinity 32.5l%o Station 26a surface temperature 57° salinity about 3 1 .4%o Off Duck Island surface temperature 56° salinity about 32. 3%o Euthemisto compressa (Plate 5) . — This common boreal amphipod occurred at twelve stations, so distributed as to show that it occurs generally over the Gulf. In IVIassachusetts Bay it was taken once (Station 44), and the records cover Piatt's Bank (Station 23), Jeffrey's Bank (Station 25), off Cape Cod (Station 43), off Cape Elizabeth (Stations 19, 22, 26b), off Seguin Island (Station 40), off Mt. Desert Rock (Station 32), the Eastern Deep Basin, both off shore and near shore (Stations 27, 36), and the neighborhood of Lurcher Shoal (Sta- tion 31). It was not taken in the closing net, and the only two cap- tures for which the horizon is known give the following data : — Station 27 surface temperature 59° salinity 32.6% Station 32 surface temperature 57° salinity 32.5% The other captures are from open hauls from intermediate depths. The largest number were taken at Station 32, surface; Station 31, at 55-0 fathoms; and Station 43, 35-0 fathoms. It was not found on German Bank (Stations 29, 30). Clione limacina (Plate 5). — Apparently this large and striking species is not abundant anywhere in the Gulf, at least in summer, though it occurs in dense swarms in the Labrador Current and be- tween Norway and Spitzbergen. Although it was taken at nine stations, Nos. 2, 6, 7, 11, 14, 19, 22, 25, 32, i. e., in Massachusetts Bay, off Cape Ann, between Jeffrey's Ledge and the coast, off Casco Bay, over Jeffrey's Bank and off Mt. Desert Rock, the total number of specimens was only sixteen, the most at any station, three; and it is such a conspicuous object in the tow, that it is not likely that any were overlooked. These stations are all near shore, the furthest out being 120 bulletin: museum of comparative zoology. only some twenty miles from land ; and so many off-shore hauls were made {e. g., Stations 23, 24, 27, 28, 29, 30, 31) that its absence from the more nearly oceanic part of the Gulf can hardly be laid to an accidental failure to capture specimens. It was not found in the northeastern part of the Gulf, nor in the Grand Manan Channel (^Stations 33, 34, 35, 36, 37) ; but its absence from the latter is probably associated with the general poverty of the zooplankton in that region. It was taken three times in the closing net, at 30 fathoms (Stations 22, 25), and 20 fathoms (Station 19), and once on the surface (Station 32). Proba- bly it would have been found oftener at the latter horizon had we made more surface hauls near shore, especially at night. But as it happened, most of the night hauls were made far off shore, where Clione was not found. The hauls afford the following data on tem- perature and salinity:^ Station 2 30 fathoms temperature 41.5° salinity 32.6%o- Station 19 20 fathoms temperature about salinity about 32.5%o. 47° Station 22 30 fathoms temperature about salinity about 32.6%o. 46.5° Station 25 30 fathoms temperature about salinity about 32.9%o. 48.5° Station 32 surface temperature 57° salinity 32.5%o. The salinity ranges from 32.5-32.9%o> the temperature from 41.5^ to 57°; and at all other stations where Clione was taken, the nets, in their course, passed tlirough waters with physical characters lying within these limits. In the Gulf, adult Clione occurs over a wide range of salinity and temperature, in water fully 10° warmer than the Labrador Current. But our collections throw no light on the condi- tions under which it reproduces. lAmacina balea (Plate 5). — The occurrence of this pteropod was even more circumscribed than that of Clione Ivmacina. It was taken at Stations 19, 22, 23, 24, 25, 30, 40, i. e., in two general regions, fu-st in the northwest corner of the Gulf, off Casco Bay and over the deep trench beyond Piatt's Bank and Jeffrey's Bank, and second, on German Bank. The known salinities and temperatures of the captures are: — Station 19 25 fathoms Station 25 30 fathoms Station 30 surface closing-net temperature 47° salinity 32 . 5%o closing net temperature salinity 32.9%o about 48° temperature salinity 32.7%o about 52° BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 121 Thus it was inhabiting rather warmer water than CHone (47°-52° as against 41*'-57°), but of about the same saUnity; and the capture at Station 30, on the surface, is particularly interesting, because Salim fusifonnis was likewise taken at that Station. The other cap- tures of Limacina were in open nets from 20-30 fathoms. T'nlike Clione, the specimens were of various ages; a swarm of small ones being taken at Station 19, the largest at the last Station at which it occurred, i. e., 40. This suggests that its chief period of grow^th is Juh' and August in the Gulf. Salpac (Plate 5). — ^ Salpae were observed over only a small area, from Station 30 to Station 31; several S. fimformi^ being taken at each Station, and others seen floating on the surface. But a large haul of S. mucronata was made twenty-five miles off Chatham, on the surface, September 30, by Capt. John McFarland of the fishing schooner Victor. The geographic importance of these hauls has been noted (p. 107). Tomopteris helgolandica. This is the only species of the genus encountered, and was taken at Stations 11, 14, 30, 32, 40, 42, and 44, i. e., in Massachusetts Bay, north of Cape Ann, off Cape Porpoise, off Mt. Desert, on German Bank, off Seguin, and once in the Kennebec River. It was not taken in any of the off-shore hauls. The known salinities and temperatures are: — Station 30 surface temperature 52° salinity S'2.7%o Station 32 surface temperature 57° salinity 32.o%o The other captures were in open nets from considerable depths (20- 60 fathoms). Chaetognaihs. — Sagittae were taken in greater or less numbers at almost every station. But the determination of most of the species of this genus is so difficult that only four, Sagitta serratodentata, S. elcgans, S. hjra, and Eukrohnia hamata have been selected, as being so easily recognized that the records can be depended upon. And the identifications of these have been verified by Mr. E. L. Michael. Sagitta serratodentata, especially, is a useful index-species, because the serrate margins of its jaAvs separate it from all its allies. Among the Sagittae in the Gra:mpus collection it is likewise readily identified by its stiff, slender body, and very large spermaries. Sagitta serratodentata (Plate 5) was taken at Stations 19, 21, 22, 23, 25, 27, 28, 30, 31, 32, 33, 36, 38, 40, 41, 44, but not in any of the bays or harbors, or in the Grand Manan Channel. The list of stations shows that it occurred very generally over the Gulf, i. e., in Massa- 122 bulletin: museum of comparative zoology. chusetts Bay, off Portland, on Piatt's and Jeffrey's Banks, over the Eastern Basin, on German Bank, and off the coast of Maine. But the table of specimens taken at each station shows that the only ones at which more than five were taken were no. 25,28, 30, 31, 32, and 36. Only one specimen was taken in Massachusetts Bay, one off Boon Island (Station 41), one off Monhegan (Station 21), and two each at Stations 19, 22, and 40. Evidently, then, its centre of abundance was off shore. It was not common anywhere near shore. The known salinities and temperatures of the captures are : — Station 25 30 fathoms closing net temperature salinity 32. 9%^ about 48° Station 27 30 fathoms closing net temperature 46° salinity 33.3%o Station 30 surface temperature 52° salinity 32. 7%o Station 32 surface temperature 57° salinity 32. 5%^ The other captures were in open nets. The largest hauls were at Stations 31 and 32, where swarms were taken. Comparison between the occurrence of S. serratodentata and S. elegans shows an interesting difference in quantitative distribution. The latter was taken at even more stations than the former, very gener- ally over the whole area, including bays and harbors. It occurred in great numbers at Stations 2 and 7, where no serratodentata were taken, and also at Station 44, where we captured only one serratodentata. Swarms of S. elegans were also encountered at Stations 12 and 14, where serratodentata w^as absent. At Stations 19, 20, 25, 27, 33, 38, 40, 44, it was numerous, from 10 to 30 or more specimens being taken at each, where serratodentata was represented by only a few specimens; and at Station 30 we encountered a swarm. On the other hand, at Stations 28, 31, 32, where we met swarms of serratodentata, they far outnumbered the elegans, as shown in the accompanying table. Number of specimens Station S. elegans S. serratodentata 2 25 0 6 2 0 7 swarm 0 11 4 0 12 swarm 0 14 20 0 19 15 2 20 30 0 BIGELOW: EX] PLORATIONS IN THE GI JLF OF MAINE. Number of specimens Station S. elegans *S. serratodentata 21 0 1 22 2 2 23 1 2 25 23 8 27 15 5 28 6 25 30 swarm 12 31 2 64 32 20 swarm 33 50 4 35 3 0 36 30 15 38 swarm 2 40 10 2 41 0 1 43 3 0 44 25 1 123 The stations at which S. elegans was most abundant (Plate 5) were 2, 7, 12, 14, 19, 20, 25, 27, 30, 32, 33, 36, 38, 44. Most of these stations are near shore; the only one which is not. Station 7, is within the influ- ence of coast water, as described above (p. 91), and the same is true of Station 25. At Station 43, however, but few were taken, and salinity shows that this is not coast water. So far as last summer's work shows, elegant is neritic in the Gulf; serratodentata oceanic. But there is, of course, no sharp line between the two. Two other chaetognaths may be mentioned here, because of their geographic importance: — Sagitta lyra, taken once, two specimens. Station 31, 55-0 fathoms, and Eukrohnia hamata, likewise taken only once, in the same haul, about twenty specimens. This species is discussed (p. 106). Medusae. — There are several Medusae of importance in the present connection. Chief among them, because so often called an Arctic form, is Staurophora mertensii; but as pointed out (p. 106) this species is not an index of polar water, for it is known from Helgoland. Large Staurophora (Plate 6) w^ere seen, and taken, at Stations 14, 15, 19, 22, 23, 25, 26, 26b, 31, 33, 34, 36, 40, 41, 42, 43, in the Grand Manan Channel, and at Eastport; showing that it was very generally distrib- uted over the Gulf, with the notable exceptions that it was not met 124 bulletin: museum of comparative zoology. with in Massachusetts Bay, at the off-shore Stations (27, 28), in the Eastern Basin nor on German Bank (Stations 29, 30). Its absence off shore is not surprising, because it is undoubtedly neritic; but its absence from Massachusetts Bay is less easily explained, because it is often very abundant there in May and June. The known salinity at which it was taken ranges from 32.5%o to 32.7%o, the known temper- ature from 50°-64°, all being surface records. But most of the actual specimens taken came from intermediate hauls with open nets; and this was notably so at Stations 14, 15, 19, 25, 36, 41, and 43, where none were taken or seen on the surface. And the Staurophorae seen floating were usually from | to 2 fathoms down, seldom on the actual surface as Aurelia so often is. None were taken in closing nets. Our records do not suggest that Staurophora is restricted to cold waters; but probably the young stages are more sensitive to temperature. Aurelia and Cyanea (Plate 6) can be considered together, as the Gulf of Maine, unlike the Norwegian Sea, has only one species, or variety, of Cyanea, which is not a migrant from elsewhere, but a per- manent inhabitant, breeding and going through its young stages here. As might have been expected, both these Medusae were most numer- ous near shore, Aurelia particularly so in the bays and harbors; and they are so large and conspicuous that they are easily seen on the surface, even if not taken in the net. In Massachusetts Bay, early in July, we saw many Aurelia, though, as it chanced, no Cyanea; but on our return thither at the end of August, both genera were seen floating on the surface at various spots between Gloucester and Provincetown. During our work along the coast between Cape Ann and Portland, the two genera were frequently recorded, both in the nets and on the surface, both of them being generally distributed in the coast waters in this region. But on the run to Piatt's Bank we left them behind at about Lat. 43° 15', long. 69° 50', and saw and took neither of them on the course thence to Jeffrey's Bank (Station 25) or until approaching the mouth of Penob- scot Bay, where (Station 26) both species once more appeared on the surface. Similarly on the run from Cape Elizabeth toward Nova Scotia the last Aurelia was seen at about 69° long. 43° 30' lat., and neither genus was found until we approached the coast again between Mt. Desert and Grand Manan. In the Grand Manan Channel, at East- port, and during the run westward along the coast, both were seen frequently, except at Stations 38 and 39. But neither species was encountered anywhere in as great abundance as they are often seen, except off Cape Cod, on August 29, when Aurelias were passed in BIGELOW: EXPLORATIONS IN THE GULF OF MAINE, 125 swarms. The greatest numl)er of Cvaneas were at Station 14 and in Penobscot Bay (Station 21a). As Damas has pointed out (Helhind Hansen and Nansen ; 1909, pt. 1, p. 101) Cyanea is one of the most important index-species of the larger plankton, because its attached stage lives in shallow water; conse- cjuently wherever Cyaneas are found ofT shore, it shows that there is a considerable admixture of coast water, and the same is true of Aurelia. Our data is important as showing that neither of them is general o\er the Gulf; both seem, if not absolutely, at least chiefly limited to a rather narrow coast-band all around the Gulf, even more so than Staurophora. And this fact suggests that there is comparatively little mixing of offshore and coast water in August. In early July as pointed out (p. 62), there is a pronounced fresh tongue oflf Cape Ann; but this flow of coast water probably reaches its maximum in June, when the Aurelias and Cyaneas are still very small, or perhaps even before they are set free. Phialidium languidum affords another example of the distribution of a neritic species. It was taken at Stations 22, 24, 25, 31, 32, 38, 40, 41, 42, 43, and in all the harbors and bays, especially Kittery, Winter Harbor and the Kennebec River; and also near Gloucester. These records show that it was found further oft" shore than either Aurelia or Cyanea, i. e., near Piatt's Bank (Station 24) and on JeflFrey's Bank (Station 25). But we did not find it on our run across the Eastern Basin toward Nova Scotia, nor on German Bank; meeting it again at Station 31 and 32, but not at Station 36. It swarmed at Station 32 and Station 40, on the surface, the salinity and temperature being : — Station 32 surface temperature 57° salinity 32.5%o Station 40 surface temperature 58° salinity about 32%o It was abundant in the harbors with lower salinity. ]Much more strictly confined to the coast water is the medusa Mclicertum campanula, which attains sexual maturity at just the time of our cruise. Great swarms were met with in Kittery Harbor, July 12 and 23, many in Gloucester and in Rockport Harbor, July 9-12; but the only outside stations at which it was taken were Nos. 4, 8, 12, 14, 22, none of them over ten miles from land. In past years, likewise, I found it very common in Penobscot Bay and at Grand Manan: but all its records in the Gulf are close to shore. Siphonophores. — Only two species of siphonophores, Agabna elegans and Physophora hydrostatica, were taken on the cruise; but their few occurrences are worth special notice because they are typical oceanic 126 bulletin: museum of comparative zoology. organisms, and both belong to the warm waters of the North Atlantic, though both are carried to Norwegian waters by the Gulf Stream. Physophora was taken at Station 30, on the surface, two speci- mens Agalma was more generally distributed, being captured at Stations 7, 27, 28, 32, 39, a total of eight very fragmentary speci- mens Unfortunately, most of them have all the organs stripped off the stem, not a tentilhim being intact; and as the latter organs are the chief generic character, identification is not beyond dispute. But the general form of the few bracts which remain attached, and of the bells taken in the same hauls, suggests identity with Agalvia elegans rather than with its close ally, Stephanomia. The records are all from the off-shore part of the Gulf. Ctenophorcs.— Two ctenophores were taken and seen frequently, Pleurohrachia pilcus and Bolinopsis scpientrionalis, neither of which was generally distributed over the Gulf, though both were taken at Pleurohrachia (Plate 6) was found at Stations 23, 27, 29, 30, 31, 32, 40 43- in the Kennebec River and off the Grand Manan Bank: several times, notably at Station 40, in great abundance. That is to say during Julv and August it was wholly absent from Massachusetts Bay, and from the coast waters between Cape Ann and Casco Bay; but was of general occurrence in the northeastern part of the Gulf over German Bank, and the Eastern Basin, as well as off <^ape Cxxl (Station 43). Swarms were encountered at Stations 30, 31, and 40; the salinities and temperatures being:— Station 30 surface temperature 52° saUnity about 32.7%^ Station 31 surface temperature 56 sa mity about 32.8%o Station 40 surface temperature 58° salmity about 32%o The salinity was not taken at Station 40, but is estimated from the records of neighboring stations. At Station 40, the swarm consisted of small individuals; at Stations 30 and 31, of large and small; and it is interesting to observe that the swarm at Station 40 was in water with very little microplankton (p. 133) while a few miles to the east, where there was an abundant microplankton, we found no Pleuro- brachia. , , n j.- j.i,„ Bolmopsis infundibulum (Plate 6) was taken (or seeii floating on the surface) at Stations 4, 6, 9, 11, 22, 25, 34, and 43, ^. . m Massachusetts Bay, the coastal waters north of Cape Ann, Jeffrey's Bank, the Grand Manan Channel, and off Cape Cod; but it was apparently absent at all the off-shore stations, at Piatt's Bank and on German Bank: nor BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. . 127 was it abundant anywhere. A third genus, Beroe, was Hkewise seen often ; and all the specimens taken belong to the cosmopolitan species B. cucumis, often recorded before from the Gulf. Results of the quantitative hauls. (Plate 7). In using the Hen sen net for quantitative hauls we were most seri- ously handicapped by working from a sailing vessel, because hauls of this sort are significant only if the vessel is practically motionless when they are taken; and it was impossible to hold the vessel motionless with the auxiliary engine in a breeze. Consequently we could carry on this line of work only at the stations which were occupied in calm weather. Small nets might have been hauled by hand from the dory at anchor; but this was not practicable with the large apparatus with which we were provided. The qualitative composition of the catches made with the Hensen net shows that they did not afford a fair esti- mate of the plankton even under favorable circumstances, because they seldom yielded any Sagittae; organisms which are plentifully represented in the four-foot net hauls. The trouble was, probably, that the nets were hauled too slowly, our hoisting engine reeling in at a rate of only about ten fathoms (about eighteen meters) per minute, which allowed the more active animals to escape. But the copepods, which usually form the bulk of the plankton of the Gulf, are more fairly represented. These shortcomings make it out of the ques- tion to draw any exact conclusion from the hauls. But they serve to show, in a general way, the relative richness of the plankton over dif- ferent parts of the region. The four-foot net hauls, too, help very materially, by supplementing the few quantitative hauls; and although I recognize that the various four-foot net hauls are not directly com- parable with one another, because rate of towing, etc. is never exactly the same at any two stations, and because the level at which the major part of the haul was made, with the open nets, might, or might not coincide with the zone richest in plankton, yet they do show, in a gen- eral way, whether the water was very rich, barren, or intermediate. And the fact that the results agree fairly well with those of the Hensen nets gives them a greater value than they could be credited with if unsupported by this more exact, though less extensive evidence. The four-foot hauls were made as nearly comparable as possible, by being of the same duration (with few exceptions \ hour) ; and by being made 128 bulletin: museum of comparative zoology. ^ith the vessel travelling at such a speed that the wire rope made an ande of about 60°, the same weight (seventy pounds) bemg invariably used The catch was placed in jars, killed with formalin, and allowed to settle, usually over night, and then measured for bulk. The Hensen net hauls were preserved entire in formalm, and measured for bulk at Cambridge, being allowed to settle before measurement, until no Lhe" visible shrinkage took place. The data of the quantitative hauls are: — Station 2 4 7 8 11 15 21 22 25 28 31 35 36 38 43 Vol. c. c. 25 5 6.5 5± 2 1 1 3 8 3 3 only a trace 3 2 1.5 Relative no. of copepods 239 104 471 30 11 97 125 25 20 trace 50 24 15 To obtain the number of copepods, the mass was diluted to 150 cc, well n^xed, and while the plankton was in suspension three cc. taken bv a pipette and counted: most of them were tried twice and the resuUs averaged. The total number of copepods in each haul is not g^en but?an be easily obtained by a simple calculation. Most of tVipm are Calanus finmarchicus. , . . ,i e TheTolumes of the four-toot qualitative hauls (om.tfng the surface hauls made with this net), in hundreds ot cub.c centimetres, are: - Station 4 6 7 8 11 12 Volume 19 19 9.5 9.5 4.7 9.5 Station 25 27 28 29 31 33 Volume 3 4.7 8 2.5 3 less than 1 BIGELOW: EXPLORATION S IN THE GULF 01 ^ MAINE. ation Volume Station Volume 14 20 34 trace 15 very small 35 trace 16 <( <( 36 7 19 4.7 38 2 20 4.7 40 3.5 21 (less than 2) 42 13 22 9.5 43 4.5 23 19 44 4.5 129 When analyzed, the foregoing tables, which in general bear each other out, show that we may separate the catches into three classes, rich, fair, and poor. The first, which I limit arbitrarily to stations where the volume of the quantitative catch was three or more cubic centimetres, and the number of copepods ninety or more in every three cc. when diluted to 150 cc. with water, includes Stations 2, 4, 7, 8, 22, and 25; the second, with quantitative hauls of one to three cc, and ten to ninety copepods, Stations 11, 15, 25, 28, 31, 36, 38, 43; the third, with quantitative haul less than one cc. in bulk, and less than fifteen copepods. Station 35. These classes agree fairly well with the volumes of the qualitative (four-foot net) hauls, as is shown by the following table, the stations in italics being the ones at which quantita- tive hauls w^ere made. 2 Qualitative 800 < 3C. ormore Qualitative 200-800 Qualitative below 200 Quantitative 3 cc. or more Quantitative 1-3 Quantitative less than 1^ with 90 or more copepods copepods 10-90 copepods, fewer than 10 2 11 16 4 15 21 "■ 6 19 33 7 20 34 8 27 35 12 28^ 14 29 22 31 23 36 ' 25 38 28 40 32 AS 42 1 Station 28 Is on the line between 1 and 2, 21 on the line between 2 and 3. 130 bulletin: museum of comparative zoology. The richest zooplankton (p. 99, 100) both in volume and in the number of copepods, was found in the northern part of Massachusetts Bay, off Cape Ann, in Ipswich Bay, over the western arm of the 100- fathom basin, off Cape Porpoise, and on Piatt's Bank; the poorest, in the Grand Manan Channel, and along the northeast coast of Maine, where the water was almost barren (p. 104). In the cold fresh water along the southern coast of Maine, and in general over the northeast- ern part of the Gulf along the west coast of Nova Scotia the richness of the plankton was intermediate, column 2 in preceding table. Along shore from Casco Bay to Penobscot Bay it was poor on our first visit early in August, but with a rich diatom plankton; and on our return, this type of plankton was found from Petit INIanan to Penobscot Ba}'; but oif the Penobscot and the Kennebec Rivers there were more copepods, enough to bring the hauls into column 2. MiCROPLANKTON. The microplankton will be the subject of a special report, conse- quently no attempt is made here to identify all the species. But its character varies so much at the different stations, and proves so char- acteristic of different regions, that the following notes are pertinent. An examination of the hauls with the no. 20 net, made at Stations 1, 6, 7, 8, 9, 12, 12a, 13, 16, 17, 19, 21, 21a, 22, 23, 24, 25, 26, 26a, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, as well as at various localities in Casco Bay, shows that the micro- plankton was of two principal types, one consisting chiefly of the peridinian, Ceratiitm tripos, the other of various diatoms, mainly species of Chaetoceras and Rhizosolenia. The diatom plankton usually contains a few Ceratium; and at several localities the two types are mixed together. Quantitatively, too, as well as qualita- tively, there is much variation between the hauls made in different parts of the Gulf (Plate 8) though our brief period of work throws no light on seasonal fluctuations. At our first few stations, in the northern half of Massachusetts Bay and in the neighborhood of Cape Ann, the microplankton proved to be very scanty in amount, consisting of a few Ceratium, an occasional Peridinium, hardly any diatoms in spite of proximity to land; but a considerable number of eggs and larvae of various Metazoa, chiefly copepods. And when we returned to Massachusetts Bay in the latter half of August, no apparent change had taken place, the hauls at BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 131 Station 44 and ofF Marshfield consisting of a few Ceratiwn tripos, with no diatoms at all, but a large amount of dirt and debris, and copepod eggs. It appears, then, that the water of the northern half of ^Massachusetts Bay, throughout July and August, w-as occupied by a very scanty Ceratium plankton, with very few diatoms. North of Cape Ann the same scanty Ceratium plankton was found occupying a belt some fifteen miles broad, as far north as Cape Por- poise, both in July (Stations 9, 11, 12b, 13) and on the return, late in August (Station 41). But in Ipswich Bay, just north of Cape Ann, close to land (Station 8), the plankton, though equally scanty, was mixed, containing a considerable number of diatoms, especially vari- ous species of Chaetoceras, and Asterionella japonica, w^hich gives it a character quite distinct from that of Massachusetts Bay, or from the neighboring stations further off shore. Xo station was occupied immediately abreast of Cape Ann on the voyage north; but on the return, August 24, we made a haul some four miles off the Cape (Station 42), finding an almost pure Ceratium plankton, with very few diatoms. But though qualitatively this agreed with ^Massachusetts Bay, it was considerably richer quantita- tively, than at any of the stations immediately north or south of the Cape. This was likewise true of our hauls over the western arm of the deep basin in early July (Station 7), and off Cape Cod at the end of August (Station 43). At both of these, Ceratiwn tripos was the prevailing organism; and with it were large numbers of Peri- dinium, but no Chaetoceras or Asterionella. Inasmuch as the samples taken at the two stations are hardly distinguishable from each other, either qualitatively or quantitatively, it is fair to assume that they represent the characteristic facies of the summer micro- plankton for the general region which they cover; one distinctly richer in mass, as well as in species, than that found in Massachusetts Baj'; but with the same organism, Ceratium tripos, occupying the leading position, and wath eciually few diatoms. The Ceratium plankton reached its maximum density over a roughly oval area southwest of Cape Elizabeth (Plate 8), which w-e traversed twice, (Stations 19, 22, 23, 26b) with an interval of seven days be- tween our two ^•isits. On our second visit, when running our line to Nova Scotia we were struck by the "slick" oily appearance of the water, some thirty-five miles off Cape Elizabeth; and consequently stopped the vessel for a surface tow (Station 26b). The net, when brought aboard, was distinctly reddish, and its meshes clogged with what proved to be a mass of Ceratium, with a \-ery few Peridiniura, 132 bulletin: museum of comparative zoology. and an occasional diatom; and this phenomenon continued for several miles. At Stations 19, 22, 23, Ceratium was not so phenomenally plentiful; but still far more abundant than at any station further to the south; and the microplankton was almost pure Ccrafium tripos, with an oc- casional Peridinium, but few, if any, diatoms; though it contained a considerable amount of copepod eggs, fish eggs, nauplii, etc., cor- responding to the rich macroplankton encountered there (p. 101). Further east, on Jeffrey's Bank, the microplankton was much less abundant, but still mostly Ccrafium tripos, with a very few diatoms, among which I noted the genus Chaetoceras, and the characteristic needle-like chains of Nitzschia scriata. Along the coast from Casco Bay to the mouth of Penobscot Bay (Stations 16, 21) there were, on the other hand, very few Ceratium, but the microplankton, which was fairly rich, in contrast Avith a very scanty macroplankton, consisted almost wholly of diatoms, the princi- pal forms being \-arious species of Chaetoceras, Thahissiosira gravida, Nitzschia scriata, and Astcrionclla japonica. Over the eastern arm of the deep basin (Stations 27, 28), the pure Ceratium plankton character- istic of the waters further west gave way to a mixed plankton, rather poor ciuantitatively, in which Ceratium was associated with a few Peridinium and \'arious diatoms, among them several species of Chaeto- ceras, and Thalassiosira gravida. And a similar type, but quantita- tively richer, was revealed by our hauls on German Bank (Stations 29, 30), where several species of Chaetoceras, Rhizosolcnia setigcra and other species of the genus, and Thalassiothrix were especially prominent in the hauls. These two stations were within a few miles of each other, and it is therefore interesting to note that at Station 29 the plankton was far richer, both quantitatively and in species, than at Station 30; and that Ceratium played, proportionately^ a greater role. However, the microplankton at both these stations, and over the eastern basin (Stations 27, 28) can be classed as Ceratium with a large admixture of diatoms, the latter probably of neritic origin for the most part. At Station 31, off Lurcher Shoal, the microplankton was very scanty, consisting chiefly of minute copepods and their eggs, and nauplii; but there were a few Ceratium and diatoms, especially Chaetoceras and Asterionella, i. c, it was of the mixed type. x\nd much the same thing was encountered off Mt. Desert Rock (Station 32), but quantitatively rather richer, the two most prominent organisms being Ceratium tripos, and the diatom Asterionella, with a few Chaetoceras BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 133 and Peridiniuin. As we approached the mouth of the Grand Manan Channel (Station 33), Ceratiuni was no longer found and the micro- plankton became very scanty, just as the macroplankton did (p. 104), consisting of \arious diatoms, chiefly Chaetoceras and Asterionella: and it grew poorer and poorer as we sailed eastward. In the Channel the microplankton was very scanty indeed, purely diatom, several species of Chaetoceras, and Asterionella l)eing the most important forms, with a few Thalassiothrix, etc. The poverty of the microplankton in the Channel was paralleled, to an even more extreme degree, by the macroplankton, and is one of the most interesting observations made on the trip, as the fact that herring occasionally swarm here shows that at times the plankton must be much more abundant than we found it. On the voyage homeward Ceratium was once more met in con- siderable numbers at Station 36, where the haul revealed a mixed plankton of the type general over the Eastern basin. (Plate 8). On August 21, when passing Great Duck Island, one of the small islands off Mt. Desert, the appearance of the water was noticeably "soupy" and immediately the vessel was hove to, and a surface haul made with the no. 20 net. When brought on board, the net was filled with a brown slimy mass which, on examination, proved to consist almost wholly of countless numbers of chains of Asterionella japonica, with a few other diatoms, particularly Chaetoceras. This phenomenon was so striking that we took frequent samples as we sailed westward, finding that the Asterionella swarm continued for some miles, though nowhere else was the mass of diatoms so dense as it was off Duck Island. At Station 38 a surface haul revealed much the same tj'pe of microplankton,. but less dense, with more Chaetoceras, and a few Peridinium, but no Ceratium. During the following night, while running from Station 38 to Station 39, a surface tow, abreast of the mouth of Penobscot Bay, was made to ascertain the extent of the Asterionella swarm; this tow revealed a diatom plankton, chiefly Asterionella, very much like Station 38, but rather less in amount. But at Station 39, we had evidently passed out of this belt, for though our hauls yielded many diatoms, there were also many Ceratium tripos; i. e. we were once more in the region of mixed microplankton; though the water was yet visibly cloudy. This phenomenon con- tinued as we crossed the mouth of Penobscot Bay, until suddenly, when some six miles ofT Seguin Island, there was a visible change and the surface water grew perfectly clear. The vessel was at once hove to, and Station 40 occupied, making a series of tows. The no. 134 bulletin: museum of comparative zoology. 20 net brought in very little indeed; but the coarser nets yielded great numbers of the common cosmopolitan ctenophore Pleurobrachia pileus, which had been previously represented only by occasional indi- viduals; thus showing that we had run out of the diatom swarm. And a pure diatom plankton was not met again on the run from the Kennebec to Cape Ann. A haul sixteen miles S. S. W. from Seguin yielded a rather barren plankton, chiefly Ceratium, with a very few Asterionella but no Chaetoceras; and, as noted above, the same type was found at Stations 41, and 42, which, with the data of stations made in July shows that a rather sparse Ceratium plankton is the normal summer type for a belt reaching from Cape Elizabeth to Cape Ann, just as it is for Massachusetts Bay. There was a striking difference between the plankton in Casco Baj^ and in Penobscot Bay. In the latter, at our only Station (21a) the water was extremely barren, there being almost no microplankton, except a few Chaetoceras. In Casco Bay (Station 16, 17, 20) on the other hand, there was an extremely rich diatom plankton, consisting almost altogether of various species of Chaetoceras and Rhizosolenia with various metazoan larvae. At Orr's Island, on July 28, the surface water was full of Chaeto- ceras and a large number of the diatom Navicula; but two days later, this type of plankton had entirely disappeared, its place being taken by hosts of ophiuran plutei, copepods, and small Medusae, e. g. Phialidium and Sarsia, without any apparent change in the physical nature of the water. BIGKLOW: EXPLORATIONS IN THE GULP OP MAINE. 135 A' C TABLE OF STATIONS. The depths are by soundings, In fathoms. 4 ft. open net. D = No. 20 silk net. H = Heiiseu quantitative not. horizontal closing net. E = Scrhn net. T = 8 ft. Beam trawl. Di- = Dredge. S = Silk net, mesh 38 per Inch. Note. To agree with the Station numbers of the U. S. Bm-eau of Fisheries 10,000 should be added to each Station number .e.g., 10,001. ■< Date Time X ^ . Nets Depths of hauls Depth of temperatuies Depth Cunent Depth H2O sample s ^ a 1 July 9 4 p. M. 42''30' 70-34' 33 TE 33,0 0,33 0,33 0,33 2 " 10 10 A. M. 42°32' 70-23' 65 AOH 0, 30, 65-0 0, 10. 35, 60 0,60 0, 40, 65 3 " " p. M. 42 "3 7' 70-22' 31 T 31 0 4 " 11 8 A. M. 42°33' 70-33' 27 AH 15-0, 31-0 0 0.27 5 " 12 3 p. M. 42°32' 70-36' 27 c. 0 0.27 0.27 0.27 6 " 13 11 A. M. 42''22' 70-43' 27 TAH. C 27, 0, 27-0, 15-0 0. 10, 27 0.27 0.27 7 " 15 4 p. M. 42°44' 69-50' 145 DAH ] 0, 75-0, 145-0 0, 25, .50, 75, 125 0. 145 0, 75. 125 8 •■ 16 8 A. M. 42<'45' 70-39' 22 HADT 22-0, 20-0, 0, 22 0,22 0.22 0.22 9 •■ 16 12 M. 42=49' 70-28' 65 D 0 0.50 0.65 0.50 10 '■ 16 4 p. M. 42''53' 70-41' 25 TA 25,0 0 11 " 17 11 A. M. 43°4' 70-20' 60 DAH 0. 30-0, 60-0 0, 15, 30, 45, GO 0,60 0, 25, 60, 80 12 " 22 2 p. M. 42''32' 70-33' 47 DA 47-0: 47-0 0 12b ■• 23 12 M. 42°53' 70-20' 80 1 0,80 0,80 0 13 " 24 12 m. 43n6' 70-20' 30 DT ' 0,30 0 14 " 24 1 p. M. 43°19' 70-13' 25 E A 0,20-0 0, 5, 15. 25 0,25 0.25 15 " 25 10 A. M. 43°37' 70- 30 T Dr. E A H. 30, 17. 0, 15-0, 20-0 0. 5, 10, 20, 30 0,30 16 " 26 11 A, M. 43°42' 69-42' 19 Dr T E A 19, 25, 0, 15-0 0, 5, 10, 15, 20 0.20 17 " 27 10 A. M. 43°41' 70-8' 15 TE 15,0 0 18 ■• 27 12 M. 43°41' 70-3' 20 Dr 20 0 19 •■ 29 11 A. M. 43°30' 69-48' 50 TE D AC. 50, 0, 0, 20-0, 25 0, 20, 30, 40, 50 0 0. eo 20 '■ 31 10 A. M. 43°39' 70-7' 10 DA 0,7-0 0 0 21 Aug. 2 3 P. M. 43°38' 69-13' 60 TADCH 60, 10-0, 0, 20, 60-0 0, 15, 30, 45, 60 0,60 0.60 21a •• 3 3 P. M. 44°5' 69-1' 8 AD 8-0,0 0 0 22 7 10 A. M. 43°26' 70-4' 47 ADHC. 30-0, 0, 45-0, 30. 0,45 45 23 7 4 P. M. 43°10' 69-40' 47 TAD 47. 20-0, 0 0 15, 25, 35, 45 0 0.45 24 7 10 P. M. 43°2' 69-19' 106 E 0 0,105 25 8 9 A. M. 43°26' 68-49' 55 DACH 0. 30-0, 30, 50-0 0, 10, 20, 30, 40, ,55 0.65 0.65 20 8 3 p. M. 43°40' 69-2' 64 D 0 0 64 0.64 0.64 26a 8 9 P. M. 43''41' 69-38' A 0 0 26b " 13 4 p. M. 43°28' 69-25' DS 0,0 0 0 27 •• 14 1 A. M. 43°26' 68-06' 100 D E A C. H. 0, 0, 80-0. 30, 90-0 0, 25, 50, 75, 100 0,100 0, 50, 100 28 " 14 9 A. M. 43°26' 67-20' 120 ADES 25-0, 0, 0, 0 0, 10, 20, 30, 60, 80, 100, 120 0 0, 30, 120 29 '• 14 6 P. M. 43°26' 66-25' 35 AED 20-0, 0, 0 0, 10, 20, 30, 35 0,35 30 •• 14 9 P. M. 43°18' 66-28' AE D 0,0,0 0 31 " 15 8 A. M. 43''45' 66-55' 75 ACDEH OO-O, 25-0, 0, 0, 70-0 0, 20, 40, 60, 75 0,75 32 '• 16 3 A. M. 43''56' 67-58' 88 AD 0,0 0,88 0.88 33 " 16 6 P. M. 44°25' 67-30' 35 AD 15-0,0 0, 5, 15, 25, 35 0.35 34 " 17 4 A. M. 44°50' 66-53' 55 AD 50-0,0 0, 10, 25, 40, 55 0, 55 35 " 19 6 P. M. 44-36' 67-11' 45 ADSH 40-0, 0, 0, 40-0 0, 10, 20, 30, 40, 45 0, 45 36 " 20 11 A. M. 44°16' 67-23' 101 ADSE 75-0, 0 0 0 0, 20, 60, 80, 100 0. 100 37 " 21 7 A. M. 44=17' 68-5' 22 AAO 22, 15-0, 0 0, 10, 20 38 " 21 6 p. M. 43°51' 68-33' 48 A.C.H. DES 40-0, 10-0, 40-0, 0, 0, 0 0, 20, 30. 40, 48 0,48 0. 48 39 " 22 9 A. M. 43-37' 69-1' 80 AD 7.5-0, 0 0, 20. 40, 60, 80 0,80 40 ■• 22 3 P. M. 43°37' 69-36' ADS 25-0, 0, 0 0 0.80 41 '• 24 11 A. M. 43''6' 70-12' ADS 20-0, 0, 0 0,80 42 " 24 9 P. M. 42°51' 70-29' ADS 15-0, 0, 0 0 43 '■ 29 12 M. 42"'ll' 69-53' 95 AHCDS 35-0, 90-0, 85-60, 0, 0 0, 20, 40, 60, 80. 95 0. 95 0. 40, 95 44 " 31 9 A. M. 42<'9' 70-22' 30 ADS 20-0, 0, 0 0. 10, 20, 30 0 0. 30 0.40 ' 0,30 45 " 31 1 P. M, 42-20' 70-36' 0, 20, 30, 40 46 " 31 3 P. M. 42-30' 70-39' 30 0,30 BIQELOW: EXPLORATIONS IN THE GULF OF MAINE. 187 TABLE OF TEMPERATURES (degrees are Fahrenheit). 5UT10!(3 1 2,3 4 ! 5 ! 6 j 7 8 9 10 U i 12 12b 13 14 15 16 17 18 10 20 55 21 21a 22 23 24 25 26 20a 26b 27 28 29 30 52 31 32 57 33 ! 61.1 50.7 34 35 36 ] 37 38 39 40 41 I42 43 44 1 45 1 46 0 5 10 15 20 25 30 35 40 45 SO 55 60 65 70 75 80 64 65 i66 56 61.3 61 64 61 67.5 61 59 60 57 57 57 55.2 58 54.5 51.4 45.5 57 53.6 51.4 47.7 45.4 57 57 57 55.5 56.3 62 64 61 56 67 57 01 50 59 150.6 56 50 51 52.5 1 55 55.6 56 58 61 61 60 68 61 61 49.2 43.3 :::::;;;;: 49.2 50.7 49.7 "1 49.7 j 49.6 .50.9 62.7 44.9 I 44.4 48.0 47.5 :::::;:::! 49.71 42.4 47.3 49.2 47.7 49.4 52.7 49.6 49.2:49.8 49.9 49. 50.9 45.9 47.7 41.1 41.3 41.3 45.3 43.3 :::: 45.2 46 49.4' 40.7 41.3 44.4 45.0 46.7 48.4 45.6 49.4 49.3 49.6 . 48.7 44.3 44.9 40.3 42.1 49.3 42.6 1 46.8 60.4 49.3 48.3 48. 4«.S 43.1 40 3 45.5 1 46.3 40.8 49.5 40.3 40. 42.2 45 ... 48.3 40.2 45.3 49.2 40.3 40.3 45 46.7 46.8 47. 42. 45.8 .... 40.3 43.3 45 8 39.4 45.3 4S.7 44.8 40.3 .... 41.S 41.3 42.8 45. 3 4S.4 ] 105 40.8 45.3 40.3 DznBB 1 ■■""1 1 BIQELOW: EXPLORATIONS IN THE GULF OP MAINE. TABLE OF SAIINITIES. STAnONB 1 2 5 6 7 8 9 1 11 12b 14 15 16 19 20 21 21a 22 23 24 25 26 27 28 29 31 32 33 j 34 35 36 38 39 41 43 44 45 46 0 6 32.07 31.74 31.67 31.96 31.62 31.44: 31.92 31.92 31.08 31.26 31.20 31.92 1 32.43 30.61 32.62 32.60 32.34 32.66 32.75 32.70 32.84 32.51 32.68 i 32.67 32.75 32.32 32.07 32.39 32.03 31.92 31.67 10 15 20 32.39 32.14 i 25 32.52 32.61 1 30 32.57 32.38 32.52 35 32.65 32.92 32.68 40 32.77 33.15 32.89 45 32.74 33.30 1 ■ 32.65 50 32.84 32.97 33.64 1 32.96 55 33.22 32.56 60 32.85 32.94 65 32.92 33.13 70 75 33.49 33.82 80 33.04 85 00 34.13 95 33.69 100 33.89 34.31 105 33.57 110 115 120 34.54 125 33.78 Dcpthi 1 Surface sample taken at this Station was lost. Boothbay Harbor. 7 fathoms, August 4. 31.71. Gloucester, surface. July 12. 31.8. Orr'a Island. July 28, 5 fathoms, 31.7. Orr's Island, surface. 31.5. Six miles S. E. Bakers Island, surface. July 15. 32.14. Gloucester, surface, July 22, 31.7. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 141 TABLE OF DENSITIES. Suiioss 1 2 5 6 1 7 S 9 11 12b 14 l.'j 16 19 20 21 21a 22 23 24 25 20 27 28 20 31 32 33 34 35 36 38 39 11 43 « 45 46 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 DmFm 23.05 22.68 23.23 23.45I 22.71 1 22.22 23.64 23.85 23.19 23.23 23.29 23.89 I 23.21 22.93 23.41 23.86 24.26 24.30 24.21 24.28 25.19 24.65 24.30 25.09 25.00 24.98 24.35 23.54 23.89 23.84 23.42 23.22 1 25.70 25.19 25.93 25.89 25.97 26.02 25.64 25.72 25.81 26.18 25.74 25.57 j 26.28 1 26.21 26.23 25.90 26.72 25.60 26.48 26.44 26.77 25.95 25.58 25.59 26.43 26.40 26.52 26.43 27.09 27.06 26.89 ■ 26.81 1 ; 27.48 27.45 27.55 27.69 28.03 i i ' Siirface sample taken at this Station was lost. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 143 TABLE OF CURRENT MEASUREMENTS. All bearings are magnetic. Surface. Stationb 1 2 4 5 6 7 8 9 lla lib lie 14 10 21 23 25 26 27 28 29 31-32 35 Plood 35 Ebb 36 37 38 41 43 44 Flow toward Velicm.persec. Knot per hour EbyS 10.2 .25 SSE 12.7 .30 EbyN 20.7 .5 NW 11.7 .25 NNE 21.0 .5 W 20.7 .5 W by S 1 NE by E 14.8 1 1.8 .37 , .05 13.2 .3 NbyW 3.3 .12 NW 17.7 .45 SW 24. .62 S 22.1 .5 NE 10.4 .25 0 0 0 NbyE 13.2 .3 NN W Trace Trace SSE 17.4 .45 BSE 12,7 .3 S .5-1. WbyS 2.-3. E 1=1= S W 1=»= NWbyW 33 .85 0 0 0 s w 8.4 .2 W SE 39.4 1 NE .78 Botiom. Stations 1 2 4 5 6 ' 8 9 lla lib 14 21 25 27 38 43 Depth fatb. 33 60 27 27 27 145 22 50 60 60 25 60 SS 100. 48 96 Flow toward EbyN 0 SE SE E E SE SSE NNW EbyN EbyS SE ESE SbyW NW NbyW Vel. cm. per sec. 10.8 0 8.7 9.7 11. 7.2 7.3 7.1 3 1.5 11.8 4.8 10.3 11.6 6.4 14. Knots per hour .25 0 .2 .25 .25 .18 .18 .18 Trace Trace .28 Trace .25 .28 .13 .37 BIGELOW: EXPLORATIONS IX THE GULF OF MAINE. 145 BIBLIOGRAPHY. Agassiz, Alexander. 1865. North American Acalephae. Mem. M. C Z., 1, 14, 234 pp., 360 figs. Apstein, Carl. 1909. Das vorkomen von Salpen in arktischen gebieten. Fauna Arctica, 5, p. 1-12, 13 figs. Bigelow, H. B. 1909. Cruise of the U. S. Fisheries Schooner "Grampus" in the Gulf Stream during July, 1908, with description of a new Medusa (Brj'tho- tiaridae). Bull. M. C. Z., 52, p. 193-210, 1 pi. 1911. Report on the scientific results of the expedition to the Eastern Tropical Pacific, 1904-1905. XXIII. The Siphonophorae. Mem. M. C. Z., 38, p. 171-401, 32 pis. 1913. Oceanographic cruises of the U. S. Fisheries Schooner "Grampus" 1912-1913. Science, 38, p. 599-601. Browne, E. T. 1910. Coelentera. V. Medusae. National Antarctic exped. Nat. hist., 5, 62 pp., 7 pis. Dawson, W. B. 1910. Effect of the wind on currents and tidal streams. Trans. Roj^al soc. Canada, ser. 3, 3, sect. 3, p. 179-196. Dickson, H. N. 1901. The circulation of the surface waters of the North Atlantic ocean. Philos. trans., ser. A, 196, p. 61-203, pi. 1-4. Ekman, V. M. 1910. Tables for sea-water under pressure. Cons. int. expl. mer., Publ. circ, no. 49, 48 pp. Esterly, C. O. 1905. The pelagic copepods of the San Diego region. Univ. Calif. Publ. zool., 2, p. 113-233, fig. 1-62. Fewkes, J. W. 1888. On certain Medusae from New England. Bull. M. C Z., 13, p. 209-240, 6 pis. Fowler, G. H. 1905. Biscayan plankton collected during a cruise of H. M . S. ' Research,' 1900. 3. The Chaetognatha. Trans. Linn. soc. London. Zool., ser. 2, 10, p. 55-88, pi. 4-7. Hautreux, A. 1910. Atlantique Nord. Bouteilles, glaces et carcasses flottantes de 1887 k 1909. Bull. Inst. oc4anog., no. 173, 18 pp., 4 charts. 1911. Temperatures de TAtlantique Nord. Bull. Inst, oceanog., no. 201, 23 pp. 146 bulletin: museum of comparative zoology. Helland-Hansen, Bjorn, and Nansen, Fridtjof. 1909. The Norwegian sea. Rept. Norwegian fish. & marine invest 2 pt. 1, 204 pp., 2, 390 pp., 28 pis. Knudsen, M. 1901. Hydrographical tables. 63 pp. Copenhagen. Kriimmel, Otto. 1907. Handbuch der ozeanographie, 1, 526 pp. Leipzig. Libbey, William. 1891. Report of a physical investigation of the waters off the southern coast of New England, made during the summer of 1889, by the U. S.. Fish Commission Schooner Grampus. Bull. U. S. fish, comm 9 p. 391-459. Mitchell, Henry. 1881. Physical hydrography of the Gulf of Maine. Rept. U. S. C. & G. S. for 1879, appendix 10, p. 175-190. Murray, John, and Hjort, Johan. 1912. The depths of the ocean. 20, 821 pp., 9 pis. London. Nordgaard, O. 1903. Studier over naturforholdene i vestlandske fjorde. I. Hydrograf. Bergens museum Aarbog, 1903, no. 8. Pettersson, Otto. 1907. On the influence of ice-melting upon oceanic circulation. Geogr. journ., 30, p. 273-295. Ritter-Zdhony, Rudolf. 1911. Revision der chaetognathen. Deutsche Siidpolar exped. Zool., 5, heft. 1, 71 pp. Schott, G. 1897. Die gewasser der Bank von Neufundland und ihrer weiten umge- bung. Petermann's mitt., 43, p. 201-212, pi. 15. 1902. Oceanographie und maritime meterologie. Wiss. ergeb. Deutsch. tiefsee-exped., 1, 40 taf. Steuer, Adolf. 1910. Planktonkunde. 16, 723 pp., 1 taf. Leipzig, etc. Tanner, Z. L. 1897. Deep-sea exploration: a general description of the Steamer Alba- tross, her appliances and methods. Bull. U. S. fish, comm., 16, p. 257-428, 40 pis. Tizard, T. H. 1907. Dr. Otto Pettersson on the influence of ice melting on oceanic cir- culation. Geogr. journ., 30, p. 339-344. Townsend, C. H. 1901. Dredging and other records of the United States Fish Commis- sion Steamer Albatross. Rept. U. S. fish. comm. for 1900, p. 387- 562, pi. 1-7. BIGELOW: EXPLORATIONS IN THE GULF OF MAINE. 147 Verrill, A. E. 1873. Results of recent dredging expeditions on the coast of New England. Amer. journ. eci., ser. 3, 5, p. 1-16, 98-106; 6, p. 435-441. Verrill, A. E. 1874. Results of recent dredging expeditions on the coast of New Eng- land, No. 6. Amer. journ. sci., ser. 3, 7, p. 405-414, pi. 4-5. 1875. Results of dredging expeditions off the New England coast in 1874. Amer. journ. sci., ser. 3, 9, p. 411-415. 1911. United States Coast Pilot. Atlantic coast, pts. 1-3. BiGELOw. — Explorations in the Gulf of Maine. EXPLANATION OF PLATES. PLATE 1. Temperatures at 25 fathoms, , and at the bottom , July and August, 1912. BiGELow. — Explorations in the Gulf of Maine PLATE 2. Salinities at the surface, , and at 25 fathoms , July and August 1912. c o X BioELOw. — Explorations in the Gulf of Maine. PLATE 3. Salinities at 50 fathoms — — — , and at the bottom , July and August, 1912. BiQBiiOW. — Explorations in tlie Gulf of Maine. PLATE 4. Circulation of water in the Gulf, July and August, 1912, as shown by salini- ties and temperatures. BiGBLOw. — Explorations in the Gulf of Maine. PLATE 5. Occurrences of pelagic animals. C. Clione limacina. L. Limacina balea. S. Sagitta serratodentata, abundant. s. " " scarce. H. Eukrohnia hamata. F. Salpa fusiformis. M. " mucronata. E. Euthemisto compressa. The curve marks the off-shore limit to abundance of Sagitta elegans, the curve the in-shore limit to abundant S. serratodentata. BiGELOW. — Explorations in the Gulf of Maine. PLATE 6. Occurrences of Medusae, Ctenophores and Siphonophores. S. Staurophora mertensii. A. Aglantha digitale. P. Pleurobrachia pileus. B. Bolinopsis infundibulum. H. Physophora hydrostatica. The curves mark the off-shore limit of abundant Aureha and Cyanea. BiGELOw. — Explorations in tiie Gulf of Maine. PLATE 7. Quantitative distribution of copepods in July and August, 1912, showing regions in which they were very abundant (1); intermediate (2); scarce (3); (see page 129). ri 6 BiGELow. — Explorations in the Gulf of Maine. PLATE 8. Distribution of microplankton, July and August, 1912. Abundant Ceratium plankton, ^p Intermediate Ceratium plankton. Scanty Ceratium plankton. V/y^ Mixed plankton. j||||| Scanty mixed plankton. : 1 1 1 1 Abundant diatom plankton. ^ Intermediate diatom plankton. Scanty diatom plankton. ^ \ BiGELOw. — Explorations in the Gulf of Maine. PLATE 9. Chart of the Gulf of Maine, with stations occupied by the Grampus, July and August, 1912, and surface temperatures. The following Publications of the Museum of Comparative Zoology are in preparation: — LOUIS CABOT. Immature State of the Odonata. Part IV. E. L. MARK. Studies on Lepidosteus, continued. " On Arachnactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II. H. L. CLARK. The "Albatross" Hawaiian Echini. with 14 Plates. Reports on the Results of Dredging Operations in 1877, 1878, 1879. and 1880, in charge of Alexander Aqassiz, by the U. S. Coast Survey Steamer "Blake," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. The Crustacea of the "Blake." A. E. VERRILL. The Alcyonaria of the "Blake." Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tanner, U. S. N., Commanding, in charge of Alexander Agassiz, as foUows: — K; BRANDT. The Sagittae. The Thalassicolae. O CARLGREN. The Actinarians. R. V. CHAMBERLIN. The Annelids. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes of Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. The Schizopods. HAROLD HEATH. Solenogaster. W. A. HE RDM AN. The Ascidians. S. J. HICKSON. The Antipathids. E. L. MARK. Branchioceriantlius. JOHN MURRAY. The Bottom Speci- mens. P. SCHIEMENZ. The Pteropods and Heteropods. THEO. STUDER. The Alcyonarians. The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Com- manding, as follows: — R. V. CHAMBERLIN. The AnneUds. H. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. S. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. The Ophiurans. <3r. W. MtJLLER. The Ostracods. MARY J. RATHBUN. The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes. T. W. VAUGHAN. The Corals, Recent and Fossil. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. There have been published of the Bulletin Vols. I. to LIV.; of the Memoiks, Vols. I. to XXIV., and also Vols. XXVI. to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, and XLI. Vols. LV. to LVIII. of the Bulletin, and Vols. XXV., XXX., XXXV., XXXIX., XL., XLII. to XLVIII. of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation : — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S. N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905, Lieut. Com- mander L. M. Garrett, U. S. N., Commanding. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. Contributions from the Geological Laboratorj', Professor R. A. Daly, in charge. These pubhcations are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology^ Cambridge, Mass. -3, V^A Bulletin of the Museum of Comparative 'Joo.'-j'is AT HARVARD COLLBOK. Vol. LVIII. No. 3. THE STANFORD EXPEDITION TO BRAZIL, 1911, JOHN C. BRANNER, Director. THE CHILOPODA OF BRAZIL. By Ralph V. Chamberlin. With Six Plates. CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM. April, 1914. Reports on the Scientific Results op the Expedition to the East- ERN Tbowc;il Vacifw, in charge of Alexander Agassiz, by the U. S. Fish Commission Steamer "Albatross," prom October, 1904, TO March, 1905, Lieutenant Commander L. M, Garrett, U. S. N. Commanding, published or in preparation: — General Report on A, AGASSIZ. V.s the Expedition. A. AGASSIZ. I.i Three Letters to Geo. M. Bowers, U, S. Fish Com. A. AGASSIZ and H. L. CLARK. The Echini. XVI." The Medusae. XXIII." TheSipho- H. B. BIGELOW. H. B. BIGELOW. nophores. H.B. BIGELOW. phores. P. BIGELOW. CARLGREN. XXVI. 2« TheCteno- R O, The Stomatopods. The Actinaria. R. V. CHAMBERLIN. The Annelids. S. F. CLARKE. VIIl.s The Hydroids. W. R. COE. The Nemerteans. L. J. COLE. XIX.n The Pycnogonida. W. H. DALL. XIV." The MoUusks. C. R. EASTMAN. VII.' The Sharks- Teeth. S, GARMAN. XII.is The Reptiles. H. J. HANSEN. The Clrripeds. H. J. HANSEN. XXVI1.2' The Schi- zopods. S. HENSHAW. The Insects. W. E. HOYLE. The Cephalopods. W. C. KENDALL and L. RADCLIFFE. XXV.26 The Fishes. C. A, KOFOID. III.3 IX.« XX.20 The Protozoa. C. A. KOFOID and J. R. MICHENER. XXII." The Protozoa. C. A. KOFOID and E. J. RIGDEN XXIV.'i The Protozoa. P. KRUMBACH. The Sagittae. R. VON LENDENFELD. XXI." The Siliceous Sponges. • The Holothurians. The Starfishes. — — The Ophiurans. G. W. MULLER. The Ostracods. JOHN MURRAY and G. V. LEE. XV] 1. 1' The Bottom Specimens. MARYJ. RATHBUN. X.>» The Crus- tacea Decapoda. HARRIET RICHARDSON. II.« The Isopods. W. E. RITTER. IV.^ The Tunicates. ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. F. E. SCHULZE. XI.u TheXenophyo- phoras. H. R. SIM ROTH. The Pteropods and Heteropods. E. C. STARKS. Xlll.ts Atelaxia. TH. STUDER. The Alcyonaria. JH. THIELE. XV." Bathysciadium. T. W. VAUGHAN. VI.« The Corals. R.WOLTERECK. XVIII.w TheAm- phipods. »Bull. M. C. Z., 2 Bull. M. C. Z.. 3 Bull. M. C. Z , * Bull. M. C. Z., » Mem. M. C. Z. 6 Bull. M. C. Z., T Bull. M. C. Z.. « Mem. M. C. Z. 'BullM. C. Z., *» Mem. M. C, Z. " Bull. M. C. Z., »2 Bull. M. C. Z., " Bull. M. C. Z., 14 Bull. M. C. Z., » Bull. M. C. Z., " Mem. M. C. Z. " Mem. M. C. Z. i« Bull. M. C. Z., '» BuU. M. C. Z., 2»Bull. M. C. Z., 2» Mem. M. C. Z. 22 Bull. M. C. Z., 2s Mem. M. C. Z. 2« Bull. M. C. Z 26 Mem. M. C. Z 26 Bull. M. C. Z. s' Mem M. C. Z., Vol. XLVL, No. 4, April, 1905, 22 pp. Vol. XLVL, No. 6, July, 1905, 4 pp.. 1 pi. Vol. XLVL, No. 9, September, 1905. 5 pp.. 1 pi. Vol. XLVL. No. 13, January, 1906, 22 pp., 3 pis. , Vol. XXXIII., January, 1906, 90 pp., 96 pis. Vol. L., No. 3, August, 1906, 14 pp., 10 pis. Vol. L., No. 4, November, 1906, 26 pp., 4 pis. ., Vol. XXXV., No. 1, February, 1907, 20 pp., 15 pis Vol. L., No. 6, February, 1907, 48 pp., 18 pis. ., Vol. XXXV, No. 2, August, 1907, 56 pp., 9 pis. Vol. LI., No. 6. November, 1907, 22 pp., 1 pi. Vol. LIL, No. 1, June, 1908, 14 pp., 1 pi. Vol. LIL, No. 2, July, 1908, 8 pp., 5 pis. Vol. XLIIL, No. 6, October, 1908, 285 pp., 22 pla. Vol. LIL, No. 5, October, 1908, 11 pp.. 2 pis. , Vol. XXXVII. , February, 1909, 243 pp., 48 pis. Vol. XXXVIII., No. 1, June, 1909, 172 pp., 5 pis., 3 maps. Vol. LIL, No. 9. June, 1909, 26 pp., 8 pis. Vol. LIL, No. 11, August, 1909. 10 pp., 3 pis. Vol. LIL, No. 13, September, 1909, 48 pp., 4 pis. , Vol. XLL, August, September, 1910, 323 pp., 56 pis. Vol. LIV., No. 7. August. 1911, 38 pp. , Vol. XXXVIII., No. 2, December, 1911, 232 pp., 32 pis. Vol. LIV., No. 10, February, 1912, 16 pp., 2 pis. ., Vol. XXXV., No. 3, April, 1912, 98 pp., 8 pis. Vol. LIV., No. 12, April. 1912, 38 pp., 2 pis. Vol. XXXV, No. 4, July, 1912, 124 pp., 12 pis. Bulletin of the Museum of Comparative ZoSIogy AT HARVARD COLLEGE. Vol. LVIII. No. 3. THE STANFORD EXPEDITION TO BRAZIL, 1911, JOHN C. BRANNER, Director. THE CHILOPODA OF BRAZIL. By Ralph V. Chambbrlin. With Six Plates. CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM. April, 1914. •^Q^ 3__ The Stanford Expedition to Brazil, 1911. John C. Branner, Director. The Chilopoda of Brazil By Ralph V. Chamberlin. The Brazilian chilopods upon a study of which the present paper is primarily a report, were collected for the most part by Mr. W. M. Mann as a member of the Stanford expedition to Brazil from June to September, 1911. As indicated hereafter, in the list by locahties and under the particular species concerned, he was assisted in certain localities by Prof. Harold Heath and in others by Dr. Fred Baker. The collection was made almost wholly in parts of Brazil from which either few or no chilopods whatsoever have been previously recorded; and its study, in connection with that of some other material from the country, has brought about such a relatively material increase in the known fauna, that it has seemed advisable to give a complete review of the chilopods of Brazil. The Stanford Expedition collection has been purchased by the Museum of Comparative Zoology. In Dr. Brolemann's Catalogue des Myriopodes du Brtsil (Sao Paulo, 1909. Catalogos de Fauna Braziliera, 2, issued by the Museu Paulista), after the elimination of manifest synonjTns and nomina nuda, there are mentioned sixteen genera and thirty-nine species of chilopods. The present paper lists seventy-one species under twenty- five genera. Of the additional forms, two families, three genera, and nineteen species have not been elsewhere recorded as occurring in Brazil, and of these one genus and seventeen species are described as new. In addition it has been deemed advisable to include descrip- tions of a new genus and three new species from the adjoining country of British Guiana, these having been studied in connection with the Brazilian material. The following list shows the known geographical distribution of the species. From states not here listed no records have been pub- lished. The new forms, and those new to the Brazilian fauna, are starred. In addition to these, because of the new territory covered, nearly all of the records of species secured by the Expedition are new within Brazil and of interest and importance in throwing light upon distribution. The greater part of pre\dously published records have been from the coastal states from Bahia southward, the most being from Bahia, Rio de Janeiro, and Sao Paulo. The States in which Mr. Mann and his associates worked are listed first and in order 152 bulletin: museum of comparative zoology. below, each being preceded by a letter; while the particular localities within the states in which collecting was carried out are indicated by a preceding number. A. Rio Grande do Norte. 1. Natal. (Mann. June). * Orphnaeus branneri, sp. nov. Scolopendra viridicornis Newport. Pselliophora nigrovittata (Meinert). Scolopopendropsis calcaratus (Pocock) . 2. Ceard-Mirim. (Mann and Heath). Orphnaeus brevilabiatus (Newport). Trematophycus celeris (Humbert and Saussure). Scolopendropsis calcaratus (Pocock). B. Ceara. 3. Ceara (Mann). Scolopendra viridicornis Newport. C. Parahyba, 4. Iridependencia (Mann and Heath. Among the hills north of the town). * Schendylurus perditus, sp. nov. * Adenoschendyla parahybae, sp. nov. Orphnaeus brevilabiatus (Newport). * Cryptops heathi, sp. nov. Pselliophora nigrovittata (Meinert). 5. Parahyba. Scolopendra morsitans Linne. D. Para. 6. Para (Mann and Baker. Chiefly in the suburb of Souza along trails through the forest. July) . Orphnaeus brevilabiatus (Newport) . * Schizonampa manni, gen. et sp. nov. * Newportia coUaris Kraepelin. * Newportia paracusis, sp. nov. Scolopocryptops miersii Newport. Otostigmus goeldi Brole- mann. Cupipes spinifer Kraepelin. Hemiscolopendra laevigata Porat. Scolopendra viridicornis Newport. Scolopendra morsitans Linn^. Santarem. Scolopendropsis bahiensis Brandt. Scolopendra gigantea Linn^. Scolopendra morsitans Linne. chamberlin: the chilopoda of brazil. 153 E. Amazonas. 7. Mandos (Mann and Baker. In and about a ruined church. August). * Schendylurus bakeri, sp. nov. Orphnaeus brevilabiatus (New- port). Notiphilides grandis Brolemann. * Mecistocephalus puncti- frons Newport. Newportia amazonica Brolemann. Newportia bicegoi Brolemann. Newportia ernsti Pocock. Newportia longitarsis (New- port). * Otostigmus amazonae, sp. nov. * Otostigmus tidius, sp. nov. Trematophycus celeris (Humbert and Saussure). Cupipes ungulatis Meinert. Cupipes ungulatis mitis Brolemann. * Cupipes amazonae, sp. nov. Scolopendra morsitans Linne. Scolopendra viridicornis Newport. 8. Porto Velho (^Slann and Baker. September). Newportia ernsti Pocock. Obidos (Brazilian Guiana). Scolopendra gigantea Linne. Carscveunc or Cal^ocnc River (Brazilian Guiana). Adenoschendyla geayi Brolemann and Ribaut. Thalthybius (Prionothalthybius) perrie'ri Brolemann. Ribautia bouvieri Brole- mann. Newportia collaris Kraepelin. F. Matto Grosso, 9. Madeira-Mamorc R. R. Camp 39. (284 km. from Porto Velho. Mann and Baker. September). NcAvportia ernsti Pocock. * Newportia longitarsis sylvae, subsp. nov. * Otostigmus rex, ?p. nov. * Otostigmus casus, sp. nov. Trematophycus celeris (Humbert and Saussure). * Cupipes neglectus, sp. nov. * Scolopendra explorans, sp. nov. 10. Madcira-Mamore R. R. Camp 4i- (On the Rio Madeira 306 km. from Porto Velho. ]\Iann and Baker. September). * Newportia longitarsis sylvae, subsp. nov. Scolopocryptops miersii Newport. * Ostostigmus suitus, sp. nov. Scolopendra anguiata Newport. Pselliophora nigrovittata (Meinert). 11. Abund. (Nearly opposite mouth of Rio Abuna. Mann and Baker. September). Trematophycus celeris (Humbert and Saussure). Corumbd. Aphilodon augustatus Silvestri. Ur Ileum. Aphilodon augustatus Silvestri. 154 bulletin: museum of comparative zoology. Pernambuco. Villa Bella. Scolopendra gigantea Linne. Pernambuco. Cupipes ungulatis Meinert. Scolopendra viridicornis Newport. Orphnaeus brevilabiatus (Newport). Rio Capivari. Cr\'ptops galatheae Meinert. Bahia. Bahia. Otostigmus scabricaudus (Humbert and Saussure). Trematophy- cus longipes (Newport). Scolopendropsis bahiensis Brandt. Sco- lopendropsis calcaratus (Pocock). Scolopendra viridicornis Newport. Scolopendra subspinipes Leach. Iguarassu. Scolopendropsis bahiensis Brandt. Rio Sao Francisco. Scolopendra viridicornis Newport. Sanio Antonio da Barra. Scolopendropsis calcaratus (Pocock). Pselliophora nigrovittata (Meinert) . MiNAS Geraes. Otostigmus scabricaudus (Humbert and Saussure). Trematophy- cus brasiliensis Kraepelin. Lagoa Santa. (?) Geophilus (Schendylurus?) sublaevis Meinert. Rio de Janeiro. Cam.po Itatiaya. Schendylurus luderwaldi Brdlemann and Ribaut. Petropolis. Cryptops iheringi Brolemann. Rio Espirito Santo. Otostigmus scabricaudus (Humbert and Saussure). Trematophy- cus brasiliensis Kraepelin. Rio de Janeiro. Orphnaeus brevilabiatus (Newport). (?) Newportia aurantiaca (Ger- vais). * Mimops occidentalis, sp. no v. Otostigmus scabricaudus (Humbert and Saussure). Cormocephalus aurantiipes (Newport). Scolopendra morsitans Linne. Scolopendra subspinipes Leach. chamberlin: the chilopoda of brazil. 155 Sao Paulo. Adenoschendyla imperfossa bolbonyx Brolemann and Ribaut. Alto da Serra. Cryptops iheringi Brolemann. Otostigmus limbatus (Meinert). Otostigmus scabricaudus (Humbert and Saussure). Bel em. Otostigmus caudatus Brolemann. Hemiscolopendra laevigata (Porat). F agenda Nova Nicaragua. ■ Schendylurus gounellei (Brolemann). Itapctininga. Otostigmus caudatus Brolemann. Piquete. Otostigmus scabricaudus (Humbert and Saussure). Otostigmus tibialis Brolemann. Poco Grande. Schendylurus paulista (Brolemann). Santos. Otostigmus limbatus Meinert. Otostigmus tibialis Brolemann. Sao Paulo. Otostigmus caudatus Brolemann. Otostigmus tibialis Brolemann, Parana. Otocrj'ptops ferrugineus macrodon Kraepelin. Otostigmus tibialis Brolemann. # Iguassu. Mecophilus neotropicus Silvestri. Santa Catherina. Blumenau. Adenoschendyla plusiodonta (Attems). Rio Grande do Sul. Cryptops iheringi Brolemann. Cr^^ptops galatheae Meinert. Porto Allegre. Scolopocryptops miersii Newport. Sao Laurengo. Scolopocryptops miersii Newport. 156 bulletin: museum of comparative zoology. Brazil (Without more definite locality). ? Schendylurus brasilianus (Silvestri). Adenoschendyla imperfossa (Brolemann). Meeistauchenius micronyx Brolemann. (?) New- portia viridis (Gervais). Otocryptops ferrugineus (Linne). Oto- cryptops melanostomus (Newport). Trematophycus longipes (New- port). Scolopendra polymorpha Wood. Scolopendra alternans Leach. Cupipes brasiliensis (Humbert and Saussure). Lithobius forficatus (Linne). It seems scarcely necessary to point out the pronounced dominance of the Scolopendroidea in the Brazilian chilopod fauna and the prac- tically complete absence of the Lithobiomorpha, so abundant in the Northern Hemisphere. It is probable, however, that a fair representa- tion of the Henicopidae will later be found to occur. In the Geo- philoidea the Schendylidae are manifestly dominant, the Oryidae coming next ; while the Geophilidae proper are at present known with certainty to be represented by but two species, one of Avhich is here first recorded and described. SCOLOPENDROIDEA. Cryptopidae. Of this family, five genera (Cryptops, Mimops, heretofore known only from China, Newportia, Otocryptops, and Scolopocryptops) are represented in the Brazilian fauna, this being about half of the total number. Cryptops Leach. Trans. Linn. Soc. London, 1814, 11, p. 384. Key to Species. a. Tarsi of all legs distinctly biarticulate; last ventral plate with scattered dark spicules or spinous points which also cover the coxopleurae C. iheringi Brolemann. aa. Tarsi of only last two pairs of legs distinctly biarticulate; last ventral plate and coxopleurae not armed with spicules, b. Second dorsal plate distinctly bisulcate; anterior margin of prosternum nearly straight C. heathi, sp. nov. bb. Second dorsal plate without sulci; anterior margin of pro- sternum distinctly biarcuate C. galatheae Meinert. chamberlin: the chilopoda of brazil. 157 Cryptops iheringi Brolemann. Rev. Museu Paiilista, 1901, 5, p. 42, pi. 1, fig. 6, 7; Kraepelin Revis. Scolop., 1903, p. 32, fig. 2; Brolemann, Cat. Myr. Br^sil, 1909, p. 8. Localities. — State of Sao Paulo : Alto da Serra (type loc.) ; State of Rio de Janeiro: Petropolis; State of Rio Grande do Sul. Cryptops galatheae Meinert. Vidensk medd. nat. foren. Kj0benhavn, 1887, p. 140; Kraepelin, Revis. Scolop., 1903, p. 54. Cryptops capivarae Pocock, Ann. mag. nat. hist., 1891, ser. 6, 8, p. 158; Brole- mann, Cat. Myr. Bresil, 1909, p. 8. Cryptops brasiliensis Attems, Mitt. Mus. Hamburg, 1900, 18, p. 112; Brole- mann, Cat. Myr. Bresil, 1909, p. 8. Localities. — State of Rio Grande do Sul (brasiliensis Attems) ; State of Matto Grosso: Rio Capivari {capivarae Pocock); Argentina: Montevideo {galatheae Meinert). Cryptops heathi, sp. nov. Color yellowish. Head a little darker than the body. Head widest anteriorly, the sides converging caudad; caudal mar- gin straight, meeting the edge of the first dorsal plate flush or scarcely overlapping the latter. Sulci not evident in middle and anterior portions; but rather faintly indicated caudad. Scarcely punctate. Prosternum not punctate; with a median longitudinal furrow. Anterior margin nearly straight, being very slightly bowed caudad toward each end; bearing three bristles on each side. (Plate 1, fig. 2). First dorsal plate with a distinct transverse cervical sulcus which is angularly bent caudad at the middle line, the plate being depressed at this angle. Two longitudinal sulci present, but these in the type are rather weak; forking cephalad, the inner branches meeting at an angle at the median line a little caudad of the cervical sulcus and the lateral lines each meeting the sulcus farther laterad (see Plate 1, fig. 1). All dorsal plates from the second to twentieth longitudinally bisul- cate. Plates not roughened; none of them bearing cornicles. Last plate with a shallow median longitudinal furrow which is more evident toward the caudal end. 158 bulletin: museum of comparative zoology. Ventral plates not punctate. Last ventral plate widely, semi- circularly, rounded caudad. Coxopleurae subtruncate caudally, not at all extended. Pores few, small, not reaching the caudal edge by a large space. Anterior legs not distinctly biarticulate. Anal legs with prefemur, femur, and, less distinctly, the tibia longitudinally furrowed dorsally. Prefemur with numerous spinules of the usual character over ventral and mesal surface; elsewhere with bristles; no longitudinal glabrous area on ventral side. Femur armed similarly to the prefemur but bearing in addition to the spinules a single stout tooth on the ventral surface toward the distal end. Tibia bearing ventrally toward the mesal edge a row of stout teeth; and the first tarsal joint bearing in corresponding location two similar teeth with an elevation distad of them as shown in Plate 1, fig. 3. Length of type cir. 10 mm. Locality. — State of Parahyba : Independencia ! (Mann and Heath) . One specimen taken in the hills north of the town. Paracryptops Pocock. Ann. mag. nat. hist., 1891, ser. 6, 7, p. 227. Kraepelin, Rev. Scolop., 1903, p. 59. Previously this genus was known only from the East Indies, from where two species had been described. Another species has been recently described from India. Paracryptops inexpectus, sp. nov. Color light lemon-yellow, darkest cephalad. Head darker than body, dilute orange-yellow excepting at anterior end where light yellow. Antennae and legs pale yellowish. Cephalic plate overlapped by the first dorsal plate. Widest toward anterior end, from where the sides at first slightly and then more strongly converge to the caudal corners; anterior border of head sub- triangular, notched at median line. On caudal portion with two short subparallel sulci; a pit-like depression a little distance from each lateral margin at middle of length; not punctate. Hair sparse. ciiamberlin: the ciiilopoda of bhazil. 159 Antennae composed of seventeen articles as usual. Densely clothed, with fine short hairs but these becoming longer and much sparser proximad. Prosternum rather widely though l)ut moderately depressed along the median longitudinal line. Anterior margin moderately extended cephalad, though less so than in hreviunguis; margin each side of the narrow median incision broadly semicircular or with edges from rounded ectal ends to mesal incision substraight, the two sides meeting at an obtuse re-entrant angle; no distinct semilunar dental plates are present though there is a suggestion of the separation of the smooth, rounded, marginal portion suggestive of the condition in weberi. First dorsal plate long; smooth; without furrows or with but very faint and short traces of a longitudinal pair toward anterior end. Sub- median paired longitudinal sulci present on other dorsal plates from the second to the penult as are also the curved lateral sulci, the latter being sharply impressed excepting on the first few plates, Avhere they are faint. Last dorsal plate with caudal produced border subtriangu- lar, the median angle somewhat obtuse with the margin each side also forming a slight, very obtuse, angle near middle of its length. Plate with a very deep median longitudinal sulcus. The ventral plates show a somewhat semicircular transverse im- pression in front of the level of the legs and a second transverse impression a little caudad of the legs, but not truely cruciform impression is indicated in the type. Last ventral plate with sides nearly straight, these converging to the semicircular caudal border. Spiracles longitudinally elliptic. Tarsi of anterior legs undivided. Legs clothed with sparse stiff bristles, these more spinescent on more caudal pairs. Prefemur of anal legs clothed with numerous long spines excepting dorsally and on median portion of ectal surface where they are replaced with fine hairs. Femur armed with similar or slightly more slender spines which are confined, however, more nearly to the strictly ventral surface; with- out teeth. Tibia without spines but bearing ventrally a longitudinal series of four teeth, these teeth curving caudad at distal ends. First tarsal joint with a single tooth at proximal end on ventral surface. Second tarsal joint without true teeth; but on ventral surface at proxi- mal end it is extended into a conspicuous rounded process. Length cir. 16 mm. Locality. — British Guiana. One specimen taken at Washington, D. C, in pots of plants imported from British Guiana. 160 bulletin: museum of comparative zoology. MiMOPS Kraepelin. Revis. Scolop., 1903, p. 62. This genus has heretofore been known from one species (M. orientalis Kraepelin) based upon a single specimen from China (Province Shensi) . It was a matter of much surprise and interest, therefore, to find in the collection of the M. C. Z. two specimens of a distinct species but fully conforming to this genus in a vial with specimens of Orph- naeus brevilabiatus (Newport) from Rio de Janeiro. The specimens of Orphnaeus had been labeled 0. brasiliensis by Meinert, who seems to have overlooked or to have failed to examine critically the smaller specimens of Mimops. It is, of course, quite possible that the speci- mens were introduced to Rio de Janeiro on ships from the East; but this must remain for the present uncertain. MiMOPS OCCIDENTALIS, Sp. UOV. The color of the types appears to have been yellowish ; but because of long preservation the original color of the specimens cannot be satisfactorily ascertained. Cephalic plate wider than long, nearly in ratio 12: 11. A little overlapping the first dorsal plate. Widest anteriorly, with sides converging caudad; caudal margin mesally a little incurved; the anterior margin incised between bases of antennae. Longitudinally depressed in caudal region each side of middle. (Plate 1, fig. 4). Antennae composed of seventeen or eighteen articles, which in the proximal half are as wide as or wider than long, but distad become longer than wide. Anterior margin of prosternum nearl}^ straight being but very weakly widely convex ; bearing two moderately high, distally rounded, dental plates separated bj' a median spaces or incision which is rounded at bottom and is deeper than in orientalis Kraepelin. All joints of prehensorial feet unarmed. Claw short and stout and but little curved. (Plate 1, fig. 5). First dorsal plate with a transverse sulcus a little caudad of margin of head, this sulcus bending caudad at middle region. Also with a longitudinal furrow each side of the middle extending cephalad from the caudal margin and uniting at an angle w^ith its fellow near the mid- dle of the plate, from where they continue as a single median furrow to the transverse sulcus (Plate 1, fig. 4). Other dorsal plates bi- chamberlin: the chilopoda of brazil. 161 sulcate. The sulci of the second and third as well as of the others entirely crossing the plate. A ridge-like elevation or keel between the sulci. Plates longitudinally depressed on each side between the sulcus and the lateral margin. (Plate 1, fig. 6). Last dorsal plate margined. Bowed out caudad with the mesal part truncate, the margin on each side of truncation slightly incurved and extending ob- liquely to the lateral margin. The two longitudinal sulci also evident on this plate excepting at caudal end. Second to nineteenth ventral plates longitudinally bisulcate. First plate with a median longitudinal furrow. Especially the more poste- rior plates longitudinally broadly depressed each side of the middle. Last ventral plate strongly narrowed caudad, though less strongly so than in orientalis. Caudal margin straight or but slightly excurved; corners rounded. Coxopleurae extended caudad in a conical process which is stouter and less cylindric than in orinitalis. Pores small and numerous. (Plate 1, fig. 7). Tarsi of all legs distinctly biarticulate. Anal legs of form very similar to that of orientalis. Prefemur with a low dorsal elevation at distal end. Claw much shorter than tarsus. Length 10.5 mm. Locality. — State of Rio de Janeiro: Rio de Janeiro. Nathaniel Thayer expedition. 1864. M. C. Z. Because of long preservation the two specimens are bleached and almost wholly bereft of hairs and spinules; accordingly, no attempt is made to describe their presence and characteristics. Mimops orien- talis Kraepelin, the other species of the genus, is from Shensi, China. It is a much larger form and differs in numerous structural details from the present species. Otocryptops Haase. Abhandl. Mus. Dresden, 1887, 5, p. 96. Scolopocryptops Newport (in part). Trans. Linn. soc. London, 1844;, 19, p. 405. Meinert (ad part, max.), Proc. Amer. philos. soc, 1886, 23, p. 179. Otocryptops Kraepelin, Revis. Scolop. 1903, p. 68. Verhoeff, Bronn's Thier- reich, 1907, 5, p. 255. Key to species. a. Twentieth, and often also the twenty-first, legs with a spine both on tibia and on tarsus; presternum with anterior margin 162 bulletin: museum of comparative zoology. bearing two to four more or less distinct teeth or dentiform elevations. ^ b. Basal tooth of first joint of prehensors small or but moderate in size, being at base from one eighth to one tenth as thick as the joint 0. ferrugineus (Linne). bb. Basal tooth of first joint of prehensors large, being at base nearly one fourth as wide as the joint. 0. ferrugineus macrodon Kraepelin. aa. Twentieth to twenty-third pairs of legs always lacking tibial and tarsal spines; prosternal margin smooth, without trace of teeth. 0. melanostomus (Newport). Otocryptops ferrugineus (Linne). Scolopendra ferruginea Linne, Syst. nat. ed., 12, 1767, 6, p. 1063. Scolopocryptops ferruginea Newport, Trans. Linn. soc. London, 1844, 19, p. 406. Scolopocryptops rufa Gervais, Insect. Apteres, 1847, 4, p. 297. Scolopocryptops mexicana Humbert et Saussure (non Saussure, 1860), Rev. mag. zool., 1869, p. 158. Scolopocryptops sexspinosa Porat (non Say), Bih. Svensk. vet. akad. Hand!., 1876, 4, no. 7, p. 26. Kohlrausch (in part). Arch, naturg., 1881, 47, 1, p. 54. Scolopocryptops antillarum Marsh, Trans. Ent. soc. London, 1878, p. 37. Scolopocryptops miersii Meinert (ad part max), Proc. Amer. philos. soc, 1886, 23, p. 181.1 Scolopocryptops hisulca Karsch, Abhandl. Naturw. ver. Bremen, 1887, 9, p. 66. Scolopocryptops strigilis Karsch, Ibid. Scolopocryptops meinerti Pocock, Ann. mag. nat. hist., 1888, ser. 6, 2, p. 474. Otocryptops ferrugineus Kraepelin, Revis. Scolop., 1903, p. 72. Otocryptops sexspinosus Brolemann (non Say, the Brazilian record). Cat. Myr. Bresil, 1909, p. 11. Locality. — " Brazil." This is a widely distributed species in Mexico, Jamaica, Haiti, and the Antilles generally, Central America, Ecuador, and Peru. 1 Of the specimens ia the M. C. Z. labeled by Meinert as S. miersii and reported upon in the paper cited above, one specimon, from Martinique, is the true S. miersii Newport, the others being O. ferrugineus. chamberlin: the chilopoda of brazil. 163 Otocryptops ferrugineus macrodon Kraepelin. Revis. Scolop., 1903, p. 74. Locality. — State of Parana: Parana (sec. Kraepelin). Separated by Kraepelin from the species on the basis of the larger tooth on the first joint of the prehensors as indicated in the key above. Otocryptops melanostomus (Newport). Scolo'pocry plops mdanosloma Newport, Trans. Linn. soc. London, LS44, 19, p. 406; Gervais, Insect. Apteres, 1847, 4, p. 298. Scolopocryptops megalocephalus Kohlrausch, Archiv. naturg., 1881, 47, 1, p. 57. Scolopocryptops luzonicus Kohlrausch, Ibid., p. 58. Scolopocryptops boholiensis Kohlrausch, Ibid., p. 58. Scolopocryptops geophilicornis Tomosvary, Termes. fiizetek, 1885, 9, p. 65. Otocryptops luzonicus Haase, Abhandl. Mus. Dresden, 1887, 5, p. 98. Otocryptops luzonicus australis Haase, Ibid., p. 98. Otocryptops longiceps Pocock, Ann. mag. nat. hist., 1890, ser. 6, 8, p. 160. Otocryptops melanostoma Pocock, Journ. Linn, soc. London, 1891, 24, p. 464; Brolemann, Ann. Soc. ent. France, 1898, 67, p. 250; Kraepelin, Revis. Scolop., 1903, p. 74, fig. 33, 34. Otocryptops aculeatus Attems, Abhandl. Senckenb. gesellsch., 1897, 23, p. 478. Locality. — Brazil (recorded as 0. longiceps by Pocock). Also known from Argentina and Venezuela and Porto Rico, St. Vincent, etc., and occurring widely in the East Indies as well. Scolopocryptops Newport. Newport (in part). Trans. Linn. soc. London, 1844, 19, p. 405; Kraepelin, Revis. Scolop., 1903, p. 76; Verhoeff, Bronn's Thierreich, 1907, 5, p. 255. The following is the only species known from the Western Hemi- sphere. Scolopocryptops miersii Newport. Trans. Linn. soc. London, 1844, 19, p. 405; Meinert (in part min.), Proc. Amer. philos. soc, 1886, 23, p. 181; Pocock, Journ. Linn. soc. London, 1893, 24, p. 146; Silvestri, Ann. Mus. civ. stor. nat. Genova, 1895, ser. 2, 14, p. 24; Brolemann, Ann. Soc. ent. France, 1898, 67, p. 250; Rev. Museu Pauhsta, 1901, 5, p. 42; Kraepelin, Revis. Scolop., 1903, p. 77; Brolemann, Cat. Myr. Br^sU, 1909, p. 33. 164 bulletin: museum of comparative zoology. Localities. — State of Matte Grosso : Madeira-Mamore R. R. camp 41 on the Rio Madeira ! (W. M. Mann) ; State of Para : Para, suburb of Souza I (Mann and Baker) ; State of Rio Grande do Sul : Porto Allegre, Sao Laurenyo. This species is common from the southern United States southward through Mexico and Central America to Venezuela, Guiana, and Brazil. Newportia Gervais. Insect. Apteres, 1847, 4, p. 298; Kraepelin, Revis. Scolop., 1903, p. 76; Ver- hoeff, Bronn's Thierreich, 1907, 5, p. 251. Newportia + Scolopendrides, Saussure, Rev. mag. zool. 1869, ser. 2, 21, p. 158. This genus, peculiar to tropical and subtropical America, is repre- sented in Brazil by seven known species, of which four have been previously recorded. Of the three here first listed from Brazil, two- are described as new. In addition, Scolopocryptops aurantiaca and S. viridis Gervais (Insect. Apteres, 4) are probably based upon members of the present genus; but there is nothing in the original descriptions to make precise identification possible, and the names must be dropped until the t^^jes are examined, if they now be in existence. Key to Species. a. Distal division of tarsus of anal legs indistinctly many ringed, the divisions not clearly separated or numerable; tibia of legs, excepting the last three pairs, armed both laterallj^ and ventrally with a stout spine; tarsus of these legs also with a stout ventral spine; spiracles very small (Scolopendrides Saussure). b. Anal leg terminating in a well-developed claw. N. amazonica Brolemann, bb. Anal leg clawless. c. Paired longitudinal sulci of head crossed near caudal ends with a fine and distinct transverse sulcus; sulci of second dorsal plate evident from anterior margin caudad to or past the middle of plate A'', ernsti Pocock. cc. Paired longitudinal sulci of head not crossed near base by any such transverse sulcus; sulci of second dorsal plate not evident on anterior half, being present only as short lines at caudal border which bifurcate into a very short mesal branch and an ectal one that runs almost directly ectad to the lateral margin A'^. paraensis, sp. no v. chamberlin: the chilopoda of brazil. 165 aa. Distal division of tarsus of anal legs composed of a limited num- ber of articles clearly separated from each other; tibia of anterior pairs of legs with only a lateral spine; tarsi of these legs without any ventral spine; spiracles large and distinct. {Nnoportia sens. str.). b. First dorsal plate with a simple, semicircular or nearly semicircular transverse cervical sulcus; its paired longi- tudinal sulcus simple and undivided; no median, pit-like, depression caudad of median sulcus. c. Prefemur of anal legs on dorsomesal surface with two rows of from seven to ten spinules. iV. longitarsis (Nevqyovt). cc. Prefemur of anal legs without any spinules proper addi- tional to the large ventral spines. N. longitarsis syhae, subsp. nov. bb. First dorsal plate with the transverse cervical sulcus bent angularly caudad at middle and with a pit-like, median, de- pression caudad of its apex; paired longitudinal sulci bi- furcating cephalad, the inner branches meeting in the depression and the lateral extending cephalo-ectad to the transverse sulcus, the branches together forming a more or less w-shaped outline. c. First joint of tarsus of anal legs clavately thickened distad and with corner drawn out at side into a pointed angle; part of first plate in front of transverse sulcus about as long as that caudad of it. A'^. collaris Kraepelin, cc. First joi^it of tarsus of anal legs not thus clavately thick- ened distad; cervical sulcus considerably cephalad of middle of first plate N. bicegoi Brolemann. Newportia amazonica Brolemann. Rev. Museu Paulista, 1903, 6, p. 69, pi. 11, fig. 3, 4; Brolemann, Cat. Myr. Bresil, 1909, p. 9. Locality. — State of Amazonas: Manaos! (Mann and Baker. This is also the type locality). Kraepelin says (Revis. Scolop., p. 85) with reference to this species: — " S. amazonica Brol. scheint sich von der vorsthenden Art [N. 166 bulletin: museum of comparative zoology. ernsti] vornehmlich durch den Besitz eines winzigen Klaue am Ende der Tarsengeissel zu unterscheiden. Ich glaube kaum das es sich hier um eine artliche Verschiedenheit, sondern um eine individuelle RiAckschlagsbildung handelt, die im interessanter Weise die Her- kunft der Newportien aus Formen mit Klauen tragenden Analbeinen demonstriert." But this author is clearly mistaken in regarding the claw thus as an individual atavistic variation, inasmuch as a normal and well-devel- oped claw is present in all the specimens secured in the type locality by Mr. Mann (Plate 2, fig. 4). In Newportia ernsti, occasionally a very small and wholly straight chitinous point may occur at the tip of the tarsus (Plate 2, fig. 5), but apparently never a true claw or any- thing that might be regarded as properly transitional to the condition in avmzonica. The spining of the prefemur, etc., is also constantly different, there being in all the specimens examined but four ventral spines instead of six, these being also relatively considerably larger. Newportia ernsti Pocock. Ann. mag. nat. hist., 1891, ser. 6, 8, p. 161; Kraepelin, Revis. Scolop., 1903, p. 8.5, fig. 38, 39; Brolemann, Cat. Mjt. Bresil, 1909, p. 10. Localities. — State of Amazonas : Manaos !, Porto Velho ! (Mann and Baker). State of Matto Grosso: Madeira-Mamore R. R. camp 39 on the Rio Madeira (W. M. Mann). Numerous specimens were secured at Manaos and several at each of the other two places indicated. Closely related to the preceding species. Newportia paraensis, sp. nov. Color yellow of a reddish tinge, with the most caudal plates darker. Head with prosternum and prehensors and the first dorsal plate reddish brown. Antennae yellow, darkest proximally. Head with punctations distinct and numerous, not fine, more sparse in frontal region, particularly in a transverse band across its caudal portion. Two median sulci distinct forward to middle of plate and not crossed by any transverse sulcus; a median longitudinal furrow from anterior margin a little distance caudad. Caudal margin widely convex, a little incurved right of middle portion of sides straight and parallel; caudal corners widely rounded. chamberlin: the chilopoda of brazil. 167 Antennae short, composed of sixteen (mostly) or seventeen articles. None of the articles glabrous but the first two more sparsely clothed with hair than the others, the density increasing from the second to the fourth and thereafter essentially uniform. Anterior margin of prosternum considerably more elevated ectally than mesally; l^earing two wide but very short dental plates which are weakly convex. First dorsal plate with a strictly semicircular cervical sulcus which is entirely free from the head and at its middle nearlv one third the distance from the head to the caudal margin of the plate. Median longitudinal sulci distinct, subparallel, crossing the transverse sulcus and attaining the front margin. Second dorsal plate with the median longitudinal sulci appearing as very short lines at caudal border, each line being continuous with a sulcus extending a little cephalad of directly ectad to the lateral margin and with a very short line extending mesocephalad. The median longitudinal sulci are complete on the other plates. The third plate shows a rather wide and shallow median longitudinal furrow. Subsequent plates with a distinct narrow median longi- tudinal keel set off by two deep furrows. Plates from fifth caudad with a distinct longitudinal furrow on each side with also less distinct indications of similar ones on the third and fourth. Last dorsal plate bowed considerably caudad, the protruding mesal portion truncate. Without a median sulcus or with but weak trace of such toward caudal end. Ventral plates from the second to the penult with a strongly marked median longitudinal furrow which is continuous from the anterior margin to a little in front of the caudal border. Lateral sulci extend- ing from anterior margin to caudal portion of plate, converging with each lateral margin. Not distinctly punctate. Last ventral plate considerably wider than long; narrowed caudad; posterior corners well rounded; caudal margin widely though but moderately convex, slightly crenately notched each side of the middle. Spiracles moderate; mostly elliptical, being obliquely or rather more dorsoventrally compressed. Coxopleural processes long and straight; ending in a single slender and acute spine; process with but scattered short hairs. Prefemur of anal legs armed ventrally with a series of six large spines and, in addition, on mesal surface with about eighteen spinules arranged in four longitudinal series. Femur armed ventrally with two spinules on proximal half and in a longitudinal line. Other joints 168 bulletin: museum of comparative zoology. unarmed. Tibia with rather sparse and moderately long hairs. Joints of tarsus with more numerous similar hairs. Second division of tarsus indistinctly segmented. First joint or division of tarsus considerably thicker than the second division and more slender than the tibia; half or a little more than half the length of the latter (17: 32.) (Plate 2, fig. 3). Length of largest specimen 28 mm. Locality. — State of Para : Para, suburb of Souza ! (Mann and Baker) . Three individuals, two adult and one immature, were secured. Newtortia longitarsis (Newport). Scolopocryptops longitarsis Newport, Trans. Linn. soc. London, 1844, 19, p. 407, pi. 40, fig. 10. Newportia longitarsis Gervais, Insect. Apteres, 1847, 4, p. 298; Humbert et Saussure, Rev. mag. zool., 1869, ser. 2, (21), p. 159; Miss, sclent. Max., 1872, p. 138; Pocock, Journ. Linn. soc. London, 1893, 24, p. 416; Brole- mann, Ann. Soc. ent. France, 1903, 67, p. 251; Kraepelin, Revls. Scolop., 1903, p. 86; Brolemann, Cat. Myr. Bresil, 1909, p. 10. Locality. — State of Amazonas : Manaos. (sc. Brolemann) ; Colom- bia; Central America, etc. Newportia longitarsis sylvae, subsp. nov. General color ochre-yellow, most of the dorsal plates being darker, more reddish, along caudal borders. Head darker, of a more ferru- ginous tinge. Antennae and legs yellowish. Head deeply and regularly, but not densely, punctate. The paired submedian sulci present only as short impressions at caudal border; a short, wider, transverse furrow, in front of their anterior ends. A rather fine median longitudinal furrow at anterior end. Antennae composed of from fifteen to seventeen articles, there being in one type specimen fifteen in the right antenna and sixteen in the left, one of those in the latter being, however, of double length and apparently representative of two normal articles. None of the articles shining or glabrous; but the hairs of the first two distinctly more sparse, those of third and fourth more dense, but only the fifth and subsequent articles fully clothed in the usual manner. Prosternum with anterior margin nearly straight, being a little more chamberlin: the chilopoda of brazil. 169 elevated at each ectal end than at middle where it is very slightly notched; edge well chitinizcd but without true dental plates. Sub- sparsely punctate. Cervical furrow of first dorsal plate strictly semicircular, distinctly exposed excepting laterally. Two median longitudinal sulci distinct; parallel excepting toward anterior ends where they diverge somewhat and finally meet the transverse sulcus; not at all branched anteriorly; but near caudal end each is joined by a strictly transverse sulcus which extends out toward lateral margin. Plate semicircularly depressed transversely in middle region near cervical sulcus (Plate 2, fig. 1). Second dorsal plate with paired sulci extending entirely across length; converging cephalad and near anterior end united with a network of very fine anastomozing transverse lines or sulci (Plate 2, fig. 1). Other dorsal plates to and including the twenty-second with two distinct and subparallel longitudinal sulci across entire length. The third plate with a fine but distinct straight sulcus running from anterior end of each longitudinal sulcus obliquely caudoectad to the lateral margin (Plate 2, fig. 1). Plates from the fifth to the twenty- second inclusive with a longitudinal furrow between each longitudinal submedian sulcus and the lateral margin. Last dorsal plate without a median longitudinal furrow. Caudally bowed out, with the ex- tended mesal portion truncate. Ventral plates with the usual median longitudinal furrow joining the distinct transverse sulcus across the caudal portion of plate but not extending across the anterior portion. Also with a distinct abbreviated longitudinal sulcus on each side convergent with lateral margin. Plates sparsely punctate. Last ventral plate considerably narrowed caudad. Caudal margin moderately incurved at middle. Wider than long in about ratio 34 : 29. Spiracles moderately large; mostly narrower ventrad, roundly subtriangular. Coxopleurae of twenty-third segment with caudal processes rather short, ending in a single spine-pore area extensive. Tarsi of anterior legs not distinctly divided. Legs clothed sparsely with stiff bristles, but with no spinules. Tibiae of anterior pairs armed laterally at distal end with a stout spine. Prefemur of anal legs armed ventrally with a row of four stout spines which increase regularly in size distad; on mesal surface with about four irregular series of small bristles and with a similar series on dorsal side near mesal edge and also similar series over ectal surface, also a few scat- tered longer bristles present, but no true spinules present. Femur 170 bulletin: museum of comparative zoology. with a stout spine on mesal surface near the proximal end and toward the ventral surface; otherwise unarmed. Other joints, so far as as- certainable from types, bearing rather scattered and short hairs, excepting the tarsi on which they are longer and more dense. Tibia broadly constricted toward each end. First article of tarsus a little more than half as long as the tibia (ratio cir. 17: 30); of same thick- ness as the immediately succeeding articles; the latter distinct and clearly separated from each other. (Plate 2, fig, 2). Length cir. 36 mm. Localities. — State of Matto Grosso : Madeira -Mamore R. R, camps 39 and 41 on the Rio Madeira! (W. M. Mann). One speci- men from each locality. Newportia collaris Kraepelin. Revis. Scolop. 1903, p. 90. Localities. — State of Para : Para, suburb of Souza ! (Mann and Baker); State of Amazonas: Lower Carsevenne, Brazilian Guiana. This second locality is the type locality and the two are the only ■ones recorded for the species. Newportia bicegoi Brolemann. Rev. Museu Paulista, 1903, 6, p. 67, pi. I, fig. 1; Kraepelin, Revis. Scolop., 1903, p. 93. Locality. — State of Amazonas : Manaos. Otostigmidae. This tropical and subtropical family is known from Australia, Asia, Africa, and the w^armer parts of America. It is represented in the known fauna of Brazil by two genera, Otostigmus, the large typical genus, and Trematophycus. Otostigmus Porat. Bih. Svensk. vet. akad. Handl., 1876, 4, no. 7, p. 18; Meinert, Vid. Medd. nat. foren. Kj0benhavn, 1884, p. 118; Proc. Amer. philos. soc, 1886, 23, p. 183; Pocock, Biol. Centr. Amer. Chilopoda, 1895, p. 25; Kraepelin, Revis. Scolop., 1903, p. 97; VerhoefT, Bronn's Thierreich, 1907, 5, p. 254, chamberlin: the chilopoda of brazil. 171 Branchiotrema Kohlrausch, Journ. Mus. Godef., 1878, p. 70; Archiv. naturg., 1881, 47, 1, p. 70. Parolostigma Pocock, Biol. Ceutr. Anier. Chilopoda, 1895, p. 25. Of this genus nine species are at present known from Brazil, five of these being here described as new. Key to Species. a. Tarsal spines wholly absent or, rarely, a few of the legs with a much reduced spine 0. limhatus (Meinert) . aa. Tarsal spines present and distinct. b. Dorsal plates of caudal half of body distinctly scabrous, bearing rows of fine elevated spinous points. c. Last ventral plate without distinct median sulcus; only the two proximal articles of the antennae glabrous; with five rather small keels or keel-like elevated lines on dorsal plates of caudal portion of the body.. .0. casus, sp. nov. cc. Last ventral plate with a distinct median longitudinal sulcus; two and a fourth or two and a half proximal articles of antennae glabrous; only a single, flat, median keel present on dorsal plates. d. Twentieth legs without a tarsal spine; only one tooth on each dental plate distinct, the others being com- pletely fused; head and first dorsal plate abruptly different in color from the other plates, brownish. 0. rex, sp. nov. dd. Twentieth legs with a tarsal spine; each, dental plate with four distinct teeth; head and first dorsal plate not abruptly different in color from the other plates, olivaceous 0. scabricaudus Humbert et Saussure. bb. All dorsal plates smooth, those of the caudal half of the body not distinctly scabrous, c. First eighteen or nineteen pairs of legs with two tarsal spines. d. First eighteen pairs of legs with two tarsal spines; twentieth legs with a tarsal spine; dorsal plates with a conspicuously elevated double keel each side of middle, the dorsal sulcus of each side lying between the halves of this keel 0. tidius, sp. nov. 172 bulletin: museum of comparative zoology. dd. First nineteen pairs of legs with two tarsal spines; twentieth legs with no tarsal spine; dorsal plates without such conspicuous keels or ridges. 0. goeldi Brolemann. cc. Legs of only the first two to the first six pairs with two tarsal spines or all with but a single tarsal spine. d. Last dorsal plate in the male ending in a process as long as the plate proper, in the female caudally acutely angular or at least rectangular; ventral plates wholly without furrows or pits 0. caudatus Brolemann. dd. Last dorsal plate in both sexes simply bowed out caudad, not acutely angular; ventral plates with distinct depressions or pits. e. Ventral plates from the second to the twentieth with distinct sulci reaching to beginning of caudal third or fourth of length; first six pairs of legs with two tarsal spines; only the first two articles of antennae glabrous 0. amazonae, sp. nov. ee. Sulci of ventral plates indicated only as short traces at the anterior border; only the first two pairs of legs with two tarsal spines; first three articles of antennae glabrous. 0. tibialis Brolemann. Otostigmus limbatus Meinert. Vid. Medd. nat. foren. Kj0behavn, 1884, p. 120; Karsch, Berl. ent. zeitschr., 1888, 32, p. 31; Silvestri, Ann. Mus. civ. stor. nat. Geneva, 1895, ser. 2, 14, p. 766; Boll. Mus. zool. anat. comp. R. univ. Torino, 1895, 10, p. 23; Brolemann, Rev. Muscu Paulista, 1901, 5, p. 37; Kraepelin, Revis. Scolop., 1903, p. 130; Cat. Myr. Bre.sil, 1909, p. 13. Localities. — "Brazil" (sec. Meinert; spec. Mus. Copenhagen); State of Sao Paulo: Alto da Serra, Santos. This species is also known from Paraguay, from where two of Meinert 's typical specimens came, and from Argentina (Buenos Aires). Otostigmus amazonae, sp. nov. Bluish green to olive-brown; with a fine median longitudinal pale line. Head distinctly darker, deeper green. Antennae bluish green chamberlin: the chilopoda of brazil. 173 proximally, paler distad. Legs more pigmented distally than proxi- mally; the posterior pairs green or bluish green excepting toward base. Head shining; pimctae weak and more or less scattered. The usual two longitudinal furrows of the caudal portion, these being shallow. i^ntennae composed of seventeen articles of which only the first two are glabrous. Prosternal teeth 4-4 of which the outermost on each side is more remote and is separated by a deeper interval than the others are from each other; innermost tooth on each side smallest, the two inter- mediate ones of nearly equal size. Dorsal plates from the fourth segment on distinctly bisulcate. Only the twenty-first plate truly margined but the others of the posterior half of the body especially, with submarginal longitudinal furrows or depressions which simulate true margination. Plates, wholly smooth and with no indication on any of a median keel. Last plate more or less angularly produced, the margin bent in on each side of the middle (Plate 3, fig. 2) ; with no developed sulci or pits, the plate somewhat longitudinally elevated along the median line and faintly depressed or furrowed along the middle of this. Ventral plates from second to twentieth with two longitudinal sulci extending to caudal third or fourth of length where each at its end is more deeply impressed. A short pit-like, median, depression in front of the caudal margin and a less pronounced median depression farther cephalad. Last ventral plate convex. Strongly narrowed caudad; caudal margin mesally excised. With a weak median sulcus and also on each side a faint fine sulcus from anterior margin to near middle (Plate 3, fig. 1). Coxopleurae not produced, being caudally simply rounded; un- armed. First six pairs of legs with two tarsal spines; the seventh to nine- teenth with one ; twentieth legs with none. Anal legs wholly unspined, being smooth throughout. Length 23 and 32 mm. Locality. — State of Amazonas: Manaos! (Mann and Baker). Two specimens were secured. This species seems to be related to 0. limbatus Meinert, but is very easily separated through the differences in the spining of the tarsi as indicated in the key. 174 bulletin: museum of comparative zoology. Otostigmus suitus, sp. nov. Color olive-green with some of the dorsal plates appearing darker along the caudal margin. Antennae more brownish excepting at base. Head distinctly punctate. Marked in front of caudal margin with two short longitudinal furrows. Antennae seventeen jointed.^ First two or two and a half articles glabrous (the third in type partially rubbed so that precise extent of glabrous condition is uncertain); other articles clothed densely with the usual short brown hairs. Dorsal plates from the fourth, inclusive, caudad with two distinct median sulci; a longitudinal depression or furrow on each side, but true margination present only on the twenty-first plate, plates mostly depressed between median sulci and wdth a weakly developed median keel indicated on plates from the third caudad; the surface and edges of plates wholly smooth. Last dorsal plate with posterior margin bowed moderately caudad. A median pit-like depression in front of caudal margin and a keel in front of this as on the other plates. (Plate 3, fig. 4). Prosternum with 4+4 teeth of which the second from the mesal incision on each side is the largest, the third being next. Ventral plates without a trace of longitudinal sulci. Each with three distinct, pit-like, or more or less longitudinal, impressions arranged in a triangle with the apex cephalad, and with three short longitudinal impressions in a transverse row in front of the caudal margin. Last ventral plate strongly narrowed caudad. Truncate caudad with the corners a little rounded. Impressed with a distinct median longi- tudinal furrow. (Plate 3, fig. 3). Coxopleurae of last legs without any true process, being a little roundly extended caudad; without any spines. Length cir. 55 mm. Locality. — State of Matto Grosso : Madeira-Mamore R. R. camp 41, on the Rio Madeira! (W. M. Mann). One specimen was secured. 1 On one side of type specimen there are but thirteen articles in the antenna, this being due, apparently, to breaking off of the antenna with subsequent regeneration of the distal article. ch.vmberlin: the chilopoda of brazil. 175 Otostigmus tidius, sp. nov. 0 Brown, of more or less ferruginous tinge caudad and also being darker ceplialad; plates mostly darker along caudal edges. Antennae very dark. Head finely densely punctate. With no true sulci; but on the anterior portion an unusually deep median longitudinal furrow and also a similar one caudad of the middle with on each side of the latter a short, more shallow, furrow diverging from it ceplialad. Antennae composed of seventeen articles of which the fu"st two are glabrous. Prosternal teeth 4+4; the three innermost on each side nearly on a level and about equal in size, but the most ectal one situated more proximad, being at about the middle of lateral edge of dental plate. Process of fu-st joint of prehensors notched or toothed on mesal side below apex. (Plate 2, fig. 6). All dorsal plates with a distinct median longitudinal furrow, on each side of which, in most of the plates, there is a double longitudinal ridge between the two edges or keels of which lies the longitudmal sulcus of the corresponding side. Ectad of tliis double keel there is a much lower, often indistinct, keel. Plates longitudinally deeply fluted or furrowed along each lateral nargin, producing the appearance of margination; but only the twenty-first plate truly margined. The keels are not well indicated on the first three or four plates. The median sulci are distinct from the fourth or fifth plates caudad. Last plate with caudal edge moderately bowed caudad. The median furrow distinct. An elevation or ridge each side of the middle divided by a weak furrow corresponding to that on the more anterior plates. (Plate 2, fig. 7). Ventral plates with indications of the longitudinal sulci over the anterior portion, but the traces very short. Without any distinct pits or similar depressions. Last ventral plate strongly narrowed caudad. Caudally convexly rounded, not at all mesally incurved. A rather fine median longitudinal furrow^ present. (Plate 2, fig. 8). Coxopleurae without true processes; but a little extended caudad, the corner being well rounded. The first to the eighteenth pairs of legs with two tarsal spines, but the lateral spine on the eighteenth minute and that of the seven- teenth intermediate in size. Nineteenth and twentieth legs with but a single tarsal spine. 176 bulletin: museum of comparative zoology. Length cir. 14.5 mm. Locality. — State of Amazonas: Manaos! (Mann and Baker). One specimen. ^ Otostigmus rex, sp. nov. Dorsum, excepting the first plate, dark olive, the plates somewhat paler along the caudal borders. Head and first dorsal plate conspicu- ously and abruptly different in color, being clear brown or somewhat testaceous, the head darker in middle region and in a narrow band running ecto-caudad on each side. Antennae and anal legs conspicu- ously rosaceous in color, the pairs of legs immediately preceding the last more weakly tinged with this color; other legs very pale clear brownish, weakly tinged with greenish. Prosternum clear brown. Venter similar to legs, darkest anteriorly. Cephalic plate punctate the punctae very fine and rather weak. Antennae composed of seventeen articles of which the first two and a half are glabrous and shining. Prosternum with each dental plate bearing a distinctly separated tooth at each ectal end; but with the other teeth thoroughly fused into a continuous plate with no or but obscure traces of the separate ones. The longitudinal sulcus between the two plates of moderate depth. Dorsal plates from the sixth on with very fine paired longitudinal sulci extending entire length of plate. From the third plate caudad there are longitudinal depressions mesad of each lateral margin which become deeper in caudal region and thus more sharply separating off the edge or simulating margination. From the third plate caudad a flat median keel is indicated, this on the anterior plates being obscure but posteriorly becoming more distinctly set off by the deepening of the limiting furrows on each side of it. Plates, especially the more caudal ones, rugose in the lateral depressions, the anterior ones other- wise smooth; but the posterior plates, and especially the last five or six, while appearing to the naked eye rather smooth, under the lens are seen to be finely scabrous, bearing over the entire surface, including edges and keel, rows of small, elevated, spinous points. Last dorsal plate bowed out caudad, the extended portion convexly rounded. In front of mesal portion of caudal edge a conspicuous and deep, pit-like, depression from the anterior edge of which a median keel runs cephalad across the plate; surface finely scabrous as on the other plates. Ventral plates without longitudinal sulci. From the third or, more chamberlin: the chilopoda of brazil. 177 indistinctly, the second to the twentieth plate with three pits, mostly deep and distinct, arranged in the form of a triangle with the median one cephalad, the three more or less clearly connected by more shallow depressions giving sometimes the appearance of a single V- or U- shaped impression. In addition there is a transverse row of three pit-like depressions in front of the caudal margin as in various related species, these pits on some of the more caudal plates lying in a more or less distinct transverse furrow. Last ventral plate long; conspicu- ously narrowed caudad; truncate or slightly inbent at the middle. A distinct longitudinal median sulcus across the entire length. Coxopleurae very slightly extended at mesocaudal corners, where they are wholly unarmed. First legs with two tarsal spines. Second to nineteenth pairs of legs with a single tarsal spine; twentieth pair with no tarsal spine. Anal legs of moderate length. Prefemur clavately widening from base distad. Wholly smooth. Length cir. 78 mm. Locality. — State of Matto Grosso: Madeira-Mamore R. R. camp, 39, on the Rio Madeira ! (Mann and Baker) . One specimen. This species lies in the group of forms closely allied with scahri- caudus in which the females are not easily distinguishable. The coloration of the present species is of a characteristic type similar to that of some Scolopendras and also present in 0. caudatus and in several African species of this genus; in these forms the head and first dorsal plate being abruptly and conspicuously different in color from the rest of the dorsum. The species also differs from scabricau- dus in being less strongly scabrous, in having no tarsal spine on the twentieth legs, in having a larger proportion of the third antennal article glabrous, and in having all the teeth excepting the most ectal on each side of the prosternum thoroughly fused together. The type is larger than the maximum measurement recorded for scabricaudus (70 mm.). Otostigmus casus, sp. nov. Olive-green in color above, brighter along the caudal margins of plates. Head more brown. Antennae brown of greenish caste, the first two articles clearer green. Legs pale brown of dilute greenish tinge. Venter lighter olive, the last ventral plate and the coxo- pleurae more brownish. Prosternum light greenish brown. 178 bulletin: museum of comparative zoology. Head subdensely punctate, the punetae being moderately fine and not sharply impressed or limited. Antennae composed of seventeen articles of which the first two are glabrous and shining, the others being densely pubescent as usual. Dorsal plates from the fifth to the twentieth with the two longi- tudinal sulci present and complete; fine. Lateral portions of plate from the fifth caudad depressed leaving the lateral margin distinctly elevated, especially in the middle and caudal regions, but true margina- tion present only on the twenty-fii'st plate. The depressed lateral portion of the plate rugose, the main rugae being longitudinal. The elevated margins, the rugae, and, less extensively, the intermediate surface, roughened with series of numerous spinulose points. From the fifth or sixth plates on a median longitudinal keel is indicated, this being at first obscure but becoming more and more distinct caudad, while at the same time on each side of it and just mesad of the sulcus appears another keel, the three keels being distinct on the caudal segments; the keels are scabrous like the lateral portions of the plates. Last dorsal plate with the posterior edge moderately bowed out caudad and mesally truncate. With three longitudinal keels corresponding to those of the other plates extending from the anterior margin caudad two thirds the length of the plate, the plate caudad of their ends having a shallow pit-like depression. Keels and general surface scabrous. Sulci of ventral plates detectable only as very short traces at the anterior border of each. With tliree pit-like depressions arranged in a triangle as usual, these being of but moderate depth and size and not coalesced. In addition there are tliree other depressions along the caudal border separated from those of the triangle by a distinct trans- verse sulcus. On some of the more caudal plates the anterior median pit may be extended a considerable distance caudad as a median furrow. Last ventral plate conspicuously narrowed caudad, the sides being convex at anterior ends but straight for most of their length. Caudal margin with lateral halves straight and meeting in the middle in a slightly reentrant angle. No distinct median sulcus present. Coxopleurae a little extended caudad at caudomesal corners which are simply rounded, no distinct process being developed, wholly unarmed. Only the first pair of legs with two tarsal spines. Second to eigh- teenth pairs with a single tarsal spine. Nineteenth to twenty-first pairs unknown, being absent from the only specimen known. Length 57 mm. chamberlin: the chilopoda of brazil. 179 Locality. — State of Matto Grosso : Madeira-Mamore R. R. camp 39, on the Rio Madeira ! (W. M. Mann) . One specimen. Otostigmus goeldi Brolemann. Ann. Soc. ent. France, 1898, 67, p. 249, pi. 20, fig. 2; Kraepelin Revis. Scolop., 1903, p. 128; Brolemann, Cat. Myr. Bresil, 1909, p. 12. Locality. — State of Para: Para (sec. Brolemann). Otostigmus scabricaudus (Humbert and Saussure). Branchiostoma scabricauda Humbert et Saussure, Rev. mag. zool., 1870, p. 203; Saussure et Humbert, Etudes Myr., 1872, p. 121, pi. 2, fig. 15, etc.; Kohlrausch, Archiv. naturg., 1881, 47, 1, p. 75. Otostigmus appendiculatus Porat, Bih. Svensk. vet. akad. Handl., 1876, 4, p. 23. Otostigma brasiliense Meinert, Vid. Medd. nat. foren, Kj0benhavn, 1884, p. 119. Otostigmus brasiliensis Karsch, Berl. ent. zeitschr., 1888, p. 31. Otostigmus scabricaudus Pocock, Ann. mag. nat. hist., 1890, ser. 6, 6, p. 142, Brolemann, Mem. Soc. zool. France, 1900, 13, p. 96; Rev. Museu Paulista, 1901, 5, p. 40; Kraepelin, Revis. Scolop., 1903, p. 126, fig. 61; Brolemann, Cat. Myr. Bre.sil, 1909, p. 13. Localities. — State of Bahia: Bahia; State of Minas Geraes; State of Rio de Janeiro: Rio de Janeiro (type locality), Rio Espirito Santo; State of Sao Paulo : Alto da Serra, Piquete. Otostigmus caudatus Brolemann. Rev. Mus. Paulista, 1901, 5, p. 39, pi. 1, fig. 1-3; Kraepelin Revis. Scolop., 1903, p. 132, fig. 71, 72; Brolemann, Cat. Myr. Bresil, 1909, p. 12. Localities. — State of Sao Paulo : Sao Paulo, Belem, Alto da Serra, Itapetininga. 180 bulletin: museum of comparative zoology. Otostigmus tibialis Brolemann. Rev. Mus. Paulista, 1901, 5, p. 39, fig. 4, 5. Otostigmus caudatus tibialis Kraepelin, Revis. Scolop., 1903, p. 132, fig. 73, 74. Otostig?7}us caiidatus Brolemann, Cat. Myr. Bresil, 1909, p. 13. Localities. — State of Sao Paulo: Sao Paulo, Piquete. Alto da Serra, Santos; State of Parana. Trematophycus Peters. Reise Mozambique, 1862, 5, p. 519. Branchiostoma Newport {norn. preocc.) Trans. Linn. soc. London, 1844, 19, p. 411; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 182. Ptychotrema Peters {nom. preocc.), Monatsb. Berl. akad., 1855, p. 82. Rhysida Wood, Journ. Acad. nat. sci. Phil., 1862, ser. 2, 5, p. 40; Pocock, Ann. mag. nat. hist., 1891, ser. 6, 7, p. 58; KraepeUn Revis. Scolop., 1903, p. 139; Verhoeff, Bronn's Thierreich, 1907, 5, p. 57. Ethmophora Pocock, Ann. mag. nat. hist., 1891, ser. 6, 7, p. 58. Of this genus three species are at present known from Brazil, and a fourth is practically certain to occur there and is accordingly intro- duced into the following key. Kep to Species. a. At least some of the dorsal plates with the paired submedian sulci crossing entire lengtlx.of plate. b. Only the twenty-first dorsal plate margined; prefemur of anal leg with one (or two) ventral spines. T. nudus (SevrpoTt). bb. Some of the dorsal plates cephalad of the twenty-first also distinctly margined. c. Prefemur of the anal legs wholly unarmed; process of coxopleurae short, subtriangular, without lateral spines. T. ceteris (Humbert and Saussure). cc. Prefemur of anal legs armed with eight to thirteen spines; process of anal coxopleurae long, with one or two lateral spines in addition to the terminal ones. T. longipes (Xe^"port). aa. None of the dorsal plates with sulci crossing its entire length, these appearing at most as short traces at ends of plates. Prefemur of anal legs wholly unspined. T. brasiliensis (Kraepelin). chamberlin: the chilopoda of brazil. 181 Trematophycus nudus (Newport). Branchiostoma nudum Newport, Trans. Linn. soc. London, LS14, 19, p. 412. Branchiosloma gymnopus Kohlraiisch, Arch, naturg., 188L 47, p. G7. Branchiostoma gymnopus ceylonicum Haasc, Abhandl. Mas. Dresden, 1887, 5, p. 86. Rhysida immarginala var. rocock, Biol. Centr. Amer. Chilopoda, 1896, p. 26. Rhijsida nuda Kraepelin, Revis. Scolop., 1903, p. 144. Locality. — While not specifically recorded from Brazil, it is dis- tributed widely in Mexico and Central America as well as in Paraguay, so that its occurrence in Brazil is practically certain. Trematophycus longipes (Newport). Branchiostoma longipes Newport, Trans. Linn. soc. London, 1844, 19, p. 411; Haase, Abhandl. Mas. Dresden, 1887, 5, p. 84. Branchiostoma obsoletum Porat. Bih. Svensk. vet. akad. Handl., 1876, 4, no. 7, p. 25. Branchiostoma gracile Kohlrausch, Archiv naturg., 1881, 47, 1, p. 06. Branchiostoma affine Kohlrausch, Ibid., p. 67. Branchiostoma longipes rotu)ulatum Haase, loc. cit. Rhysida longipes Pocock, Biol. Centr. Amer. Chilopoda, 1896, p. 27; Kraepelin, Revis. Scolop., 1903, p. 148, fig. 91; Brolemann, Cat. Myr. Bresil, 1909, p. 14. Localities.— "BraziV (M. C. Z. coll.; also W. M. Mann, without precise locality) ; State of Bahia. Trematophycus celeris (Humbert and Saussure). Branchiostoma celer Humbert et Saussure, Rev. mag. zool., 1876, ser. 2, 22, p. 202; Kohlrausch, Arch, naturg., 1881, 47, 1, p. 69, Meinert, Proc. Amer. philos. soc, 1886, 23, p. 183. Rhysida celeris Silvestri, Mus. zool. anat. comp. R. univ. Torino, 1895, 10, p. 23; Pocock, Biol. Centr. Amer. Chilopoda, 1896, p. 27; Kraepelin, Revis. Scolop., 1903, p. 149; Brolemann, Cat. Myr. Br6sil, 1909, p. 14. Localities. — State of Matto Grosso: Madeira-Mamore R. R, camp 39, on the Rio Madeira!, Abuna, Bolivia! (W. M. Mann); State of Rio Grande do Norte: Ceara-Mirim! (Mann and Heath. The specimen from this locality variant) . 182 bulletin: museum of comparative zoology. Trematophycus brasiliensis (Kraepelin). Rhysida brasiliensis Kraepelin, Revis. Scolop., 1903, p. 152, fig. 95, 96. Localities. — State of Minas Geraes ; State of Rio de Janeiro : Rio Espirito Santo. Scolopendridae. Of this family the genera Scolopendropsis, Cupipes, Cormocephalus (probably introduced), Hemiscolopendra, and Scolopendra are known in the Brazilian fauna. Scolopendropsis Brandt. Bull. sci. Acad, imper. sci. St. Peterburg, 1840, 7, p. 24; Kraepelin Revis. Scolop., 1903, p. 179; Verhoeff, Bronn's Thierreich, 1907, 5, p. 50. Rhoda Meinert, Proc. Amer. philos. soc, 1886, 23, p. 188. Pithopus Pocock, Ann. mag. nat. hist., 1891, ser. 6, 7, p. 223; Kraepelin, Revis. Scolop., 1903, p. 171; Verhoeff, Bronn's Thierreich, 1907, 5, p. 259. Two species, the only valid ones known, occur in Brazil. Scolopendropsis BAHiENSis^Brandt, Bull. sci. Acad, imper. sci. St. Peterburg, 1840, 7, p. 24; Kraepelin, Revis. Scolop., 1903, p. 171; Brolemann, Cat. Myr. Bresil, 1909, p. 31. Rhoda thayeri Meinert, Proc. Amer. philos. soc, 1886, 23, p. 188. Pithopus inermis Pocock, Ann. mag. nat. hist., 1891, ser. 6, 7, p. 223, pi. 5, f. 5; Kraepelin, Revis. Scolop., 1903, p. 172. Scolopendropsis thayeri Brolemann, Cat. Myr. Bresil, 1909, p. 32. Localities. — State of Bahia : Bahia, Iguarassu (the type locality of inermis (Pocock)). State of Para: Santarem (the type locality of thayeri (Mein.)). Kraepelin suggests the identity of thayeri Meinert with calcaratus Pocock; but an examination of the type of thayeri shows that it is the same as the inermis of Pocock and that both are the same as Brandt's hahiensis Avhich has long priority. _ Through probable error twenty- three pairs of legs were attributed to Brandt's species, although the not very probable suggestion has been made that the species is di- morphic, having some individuals with twenty-three and others with twenty-one pairs of legs. chamberlin: the chilopoda of brazil. 183 SCOLOPENDROPSIS CALCARATUS (Pocock). Pithopus calcaralus Pocock, Ann. mag. nat. hist., 1891, ser. 6, 7, p. 224, pi. 5, fig. 3. Scolopendropis calcaralus Brolemann, Ann. Soc. ent. France, 1902, 71, p. 651; Cat. Myr. Bresil, 1909, p. 32. Localities. — State of Bahia: Bahia (t\TDe locality), Santo Antonio da Barra; State of Rio Grande do Norte: Natal! (W. M. Mann). Ceara-Mirim I (Mann and Heath). CupiPES Kohlrausch. Archiv. naturg. 1881, 47, 1, p. 78; Kraepelin, Revis. Scolop., 1903, p. 174; Verhoeff, Bronn's Thierreich, 1907, 5, p. 260. Key to Species. a. Femur or second joint of anal legs armed dorsally at distal end with two bifid spines, one on each side of the longitudinal furrow. C. hrasiliensis Humbert and Saussure. aa. Femur of anal legs unarmed. b. Only the twenty-first dorsal plate distinctly margined. c. Most legs with a tarsal spine; last ventral plate parallel sided, not narrowed caudad C. spinifer Kraepelin. cc. All legs without tarsal spines; last ventral plate narrowed caudad. d. Prefemur of anal legs with about five to eleven spines distributed over its surface . . . . C. ungulatis Meinert. dd. Prefemur with only two spines, these being those at dorsomesal angle of distal end. C. ungulatis mitis Brolemann. bb. Plates from seventh to tenth caudad also margined as well as the twenty-first. c. Coxopleura with a short but distinct process at meso- caudal corner (Plate 2, fig. 9); each dental plate of prosternum with an elongated ectal process and a single mesal tooth at base of this (Plate 3, fig. 5); head dis- tinctly longer than wide C. amazonae, sp. nov. cc. Coxopleura with no such process, simply rounded at meso- caudal angle (Plate 3, fig. 8) ; each dental plate with four subsimilar teeth of the usual form (Plate 3, fig. 7) ; head equal in length and breadth C. neglectus, sp. nov. 184 bulletin: museum of comparative zoology. CuPiPES BRASiLiENSis (Humbert and Saussure). Cormocephalus brasiliensis Humbert et Saussure, Rev. mag. zool., 1870, ser. 2, 22, p. 203; Saussure et Humbert, Etude Myr. 1872, p. 124, pi. 6, fig. 17. Cupipes brasiliensis Brolemann, Cat. Myr. Bresil, 1909, p. 8. Locality. — Brazil. This species was regarded, doubtfully, as the same as C. ungulatis Meinert by Meinert as well as by Kraepelin. But, as pointed out by Brolemann, and as indicated in the key above, the presence of the dorsal spines on the second joint of the anal legs seems clearly to separate hrasiliensis from Meinert's species. At any rate, until the Brazilian fauna is better known, or specimens of ungulatis are shown to present the important variation mentioned, it would be quite premature to ignore this characteristic. Cupipes spinifer Kraepelin. Revis. Scolop., 1903, p. 177, fig. 117. Locality. — State of Para : Para (type locality) . Known from a single specimen in the Hamburg Museum. Cupipes ungulatis Meinert. Proc. Amer. Pliilos. soc, 1886, 23, p. 187; Brolemann, Rev. Mu.seu Paulista, 1903, 6, p. 64; Cat. Myr. Bre.sil, 1909, p. 9. Localities. — State of Pernambuco : Pernambuco ! (type locality) ; State of x\mazonas: Manaos (sec. Brolemann). Cupipes ungulatis mitis Brolemann. Rev. Museu Paulista, 1903, 6, p. 65; Cat. Myr. Bresil, 1909, p. 9. Locality. — State of Amazonas: Manaos (type locality). This and the preceding form are known at present from too few specimens to be able properly to judge of their precise relationship. Considerable variation has already been noted in the spining of the prefemur of C. ungulatis; and it may prove not possible to segregate the forms on the basis of this character. CHAMBERLl.N : THE CHILOPODA OF BRAZIL. 185 CUPIPES .^AlAZONAE, Sp. nOV. Dorsum olive or oli\e-broAvn; most of the plates of the middle and anterior regions of body with a blackish spot or short stripe at each lateral margin; in the plates of the median region also a dark spot on caudal border at caudal end of each sulcus or the two spots may be united as a transverse band. Head darker, more brownish and dusky. Antennae bright green. Prosternum and prehensorial feet clear brown, the latter rufous laterally and especially distally proximad of the black claw proper. Head clearly longer than wide (46:41). Sides a little convex just behind the eyes and then substraight and a little converging to the caudal corners. Finely and uniformly subdensely punctate. The two di^■erging longitudinal sulci reaching or very nearly reaching the anterior margin, each terminating in a transverse sulcus slightly removed from the edge of antennal socket with which it runs parallel. Antennae composed of seventeen articles of which the first seven are glabrous or practically so and in this respect sharply separated from the more distal group. Prosternum with two sharply defined longitudinal sulci converging cephalad and uniting at an angle at anterior end; these sulci crossed by a transverse sulcus which is branched and has anastamoses as shown in Plate 3, fig. 5. Dental plate without true teeth, but bearing an elongate ectal process with on mesal side at base a low, dark, denti- form elevation (Plate 3, fig. 5); on one side the plate is malformed as shown in the figure, this probably being due to injury with imper- fect subsequent regeneration. IMargination of the dorsal plates indicated from about the seventh caudad but not very distinctly until the tenth. Plates with an obscure low median keel defined by two indistinct longitudinal furrows; also a vague furrow laterad of each median sulcus may be indicated. Last dorsal plate with a sharply impressed median longitudinal sulcus. Second to twentieth ventral plates with the usual two longitudinal sulci crossing the plate. A rather vague transverse furrow may be traced at the level of the legs; while on some plates indications of a verv weak median longitudinal furrow mav be detected. Last ven- tral plate clearly wider at its anterior end than long (2.6:2). Sides strongly converging caudad; plate truncate caudad, the corners a little rounded. \Yithout a median sulcus or furrow. (Plate 2, fig. 9). Coxopleurae with a short but distinct process at mesocaudal corner, 186 bulletin: museum of comparative zoology. the process bearing two spines; not spined on coxopletiral margin laterad of the process proper. Porose area not fully reaching the caudal margin. (Plate 2, fig. 9). Anal legs with articles proximad of the tarsus much thickened, especially from side to side as usual. Prefemur and especially the femur deeply longitudinally furrowed dorsally at distal ends. Pre- femur with a distinct spine at mesodistal angle on dorsal side; two small spines near upper edge of mesal surface and three more ventral, two being at the distal end ventrad of the corner spine and one toward the proximal end; in addition there are four strictly ventral spines, two in each of two rows. Femur wholly unarmed. Length cir. 43 mm. Locality. — State of Amazonas: Manaos! (Mann and Baker). One specimen. CupiPES neglectus, sp. nov. Dorsum brown. Head olivaceous. Antennae bluish green. Legs pale. Head equal in length and breadth. Sulci distinct, diverging cepha- lad and each reaching the margin at the eye. Punctae scattered. Antennae composed of seventeen articles. In the type the antennae are considerably rubbed; but apparently the first article is wholly glabrous and the second one nearly so, the third and fourth with an intermediate number of hairs and the fifth* and subsequent ones with the full complement. Prosternum with two longitudinal submedian sulci which meet at an angle anteriorly; not crossed by any distinct transverse lines. Dental plates bearing 4+4 teeth which are of normal form and long and acute; the two intermediate teeth on each side longest; the most eetal tooth situated distinctly more proximad (Plate 3, fig. 7). Dorsal plates margined from the eighth or ninth segment caudad, the margination becoming more and more distinct in going toward the caudal end. Sulci continuous and very distinct on all the plates excepting the last. First to third plates with a median longitudinal furrow; the fourth plane; those from the fifth caudad with a low me- dian longitudinal keel which is flat and set off by two shallow furrows in the usual way. Last dorsal plate with a sharply impressed median longitudinal sulcus. Ventral plates from the second to the twentieth distinctly bisulcate. Some of the plates showing an indistinct median depression on caudal chamberlin: the chilopoda of brazil. 187 portion. Last ventral plate narrowed caudad; sides straight; caudal margin also straight. A little wider across anterior end than long (ratio cir. 39:37). With a rather weak median longitudinal sulcus. Coxopleurae of last legs not at all produced caudad. Bearing one or two closely approximate spines at mesocaudal angle ; none ectad of this. Porose area not reaching caudal margin by a considerable distance. Length of the two types 26 and 55 mm. respectively. Joints of the anal legs proximad of the tarsus strongly crassate as usual. The first three joints longitudinally dorsally sulcate, the sulci most distinct at distal ends and on the first two joints. Spines of prefemur very small and difficult to detect; two present close to- gether on dorsomesal corner of distal end; a single one in line with the preceding two farther cephalad and three on the mesal surface were detected. First twenty pairs of legs without tarsal spines. Locality. — State of Matto Grosso : Madeira-Mamore R. R. camp 39, on the Rio Madeira! (W. M. Mann). Two specimens. Evidently this species is related to C. impressus (Porat); but it may be distinguished in having the cephalic plate equal in length and breadth instead of distinctly longer (ratio 4:3), by having the last ventral plate with a median longitudinal furrow, by having fewer articles of the antennae glabrous and these less abruptly differentiated from the others, etc. CoRMOCEPHALUS Newport. Trans. Linn. soc. London, 1844, 19, p. 275, 419; Meinert, Proc. Amer. philos. see, 1886, 23, p. 205; Kraepelin, Revis. Scolop., 1903, p. 184; Verhoeff, Bronn's Thierreich, 1907, 5, p. 262. Rhombocephalus Newport (in part) Trans. Linn. soc. London, 1844, 19, p. 275, 425. This is a typically Australian, Indian, and African genus, the single species recorded from America having probably been introduced from ships. CoRMOCEPHALUS AURANTiiPES (Newport). Scolopendra aurantiipes Newport, Ann. mag. nat. hist., 1844, 13, p. 99. Scolopendra subminiata Newport, Ibid., p. 100. Cormocephalus aurantiipes 'Newport, Trans. Linn. soc. London, 1844, 19, p. 420; 188 bulletin: museum of comparative zoology. Haase, Abhandl. Mus. Dresden, 1887, 5, p. 57; Kraepelin, Revis. Scolop,, 1903, p. 197; Brolemann, Cat. Myr. Bresil, 1909, p. 7. Cormocephalus obscwus Newport, Loc. cit., p. 420. Cormocephalus subminiatus Newport, Loc. cit., p. 420. Cormocephalus miniatus Newport, Loc. cit., p. 420. Rhombocephalus hrevis Newport, Loc. cit., p. 426. Scolopendra obscura Gervais, Insect. Apteres, 1847, 4, p. 272. Scolopendra niiniata Gervais, Ibid. Scolopendra subminiata Gervais, Ibid. Scolopendra brens Gervais, Ibid. Cormocephalus gracilis Kohlrausch, Ai'chiv. naturg., 1881, 47, 1, p. 86. Cormocephalus pygometas Kohlrausch, Ibid., p. 90. Cormocephalus aurantiipes spinosus Haase, Loc. cit., p. 58. Localities. — State of Rio de Janeiro: Rio de Janeiro! M. C. Z.; Brazil (without more definite locality, recorded as C. gracilis and C. pygomelas by Kohlrausch.) Reported also from Guatemala by Pocock. HeMiscolopendra Kraepelin. Revis. Scolop., 1903, p. 212; Verhoeff, Bronn's Thierreich, 1907, 5, p. 261. Hemiscolopendra laevigata (Porat). Cormocephalus laevigatus Porat, Bih. Svensk. vet. akad. Handl., 1876, 4, no. 7, p. 17. Scolopendra cormocephalina Kohlrausch, Archiv. naturg., 1881, 47, 1, p. 123. Scolopendra longispina Meinert, Proc. Amer. philos. soc, 1886, 23, p. 199; Brolemann, Cat. Myr. Bresil, 1909, p. 19. Scolopendra appendiculata Daday, Termes, fiizetek., 1891, 14, p. 152. Localities. — State of Sao Paulo: Belem. Scolopendra Linne. Syst. nat. ed. 10, 1758, 4, p. 637; Newport, Trans. Linn. soc. London, 1844, 19, p. 275, 377; Latzel, Myr. Ost.-Ung. monarch., 1880, 1, p. 137; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 190; Vid. Meddl. nat. foren. Kj0benhavn, 1886, p. 125; Kraepehn, Revis. Scolop., 1903, p. 223; Ver- hoeff, Bronn's Thierreich., 1907, 5, p. 263. Eight valid species, one of them new, are recognized in this paper as occurring in Brazil. Brolemann lists S. viridis Wood as occurring CHAMBERLIN: THE CHILOPODA OF BRAZIL. 189 in Brazil, fide Meinert; but a reference to the paper cited fails to reveal any record of the species from the country by the latter author and the species is accordingly here omitted. The names falling as synonyms in this genus are very numerous. The known Brazilian species may be separated by means of the following key. Key to Species. First dorsal plate with a deeply impressed transverse cervical sulcus which may be nearly covered by the head, b. Prefemur of the twentieth legs (as also sometimes of some of the immediately preceding pairs) armed on dorsal side at distal end one to several spines; prosternum with a fine sul- cus across anterior portion. c. Dorsal plates from the sixth or seventh caudad distinctly margined laterally. d. At least eight or ten basal articles of the antennae glabrous, the others finely pubescent, the hairs not in rows; coxopleural process with nine spines or points S. (jigantea Linne. dd. Only four or five basal articles of antennae glabrous; hairs of the others often in streaks or rows; points of the coxopleural process fewer. e. Ventral plates without longitudinal sulci ; margina- tion of dorsal plates beginning at seventh segment; femur of twentieth and also of nineteenth legs with one or two spines at distal end above. S. angulata Newport, ee. Ventral plates of second to twentieth segments with two deep longitudinal sulci crossing the entire length of plate; margination of dorsal plates be- ginning with the fifth; femur of twentieth legs lacking spine at distal end above. S. viridicornis Newport, cc. Margination of dorsal plates beginning only with the eleventh to fourteenth segment. Four basal articles of antennae glabrous, the others densely pubescent; none of dorsal plates with sulci passing across entire length; claw of anal legs with two minute basal spines S. explorans, sp. nov. 190 bulletin: museum of comparative zoology. bb. Prefemur of twentieth legs not at all armed at distal end above; prosternum with no transverse sulcus across entire width of anterior portion. Head wholly without longitudinal sulci; last dorsal plate with a median longitudinal sulcus; antennae composed of from twenty-five to thirty-one articles of which the eighth to seventeenth proximal ones are glabrous. S. liolymorpha Wood, aa. First dorsal plate without any such deeply impressed transverse cervical sulcus. b. Prefemur of nineteenth and twentieth pairs of legs armed dorsally at distal end with from one to six spinous teeth; head with longitudinal sulci S. alternans Leach. bb. Prefemur of nineteenth and twentieth pairs of legs unarmed at distal end above; head without longitudinal sulci. c. Prefemur of anal legs armed ventrally with from six to nine spines; last dorsal plate with a fine median longi- tudinal sulcus S. morsitans Linne. cc. Prefemur of anal legs with only three, or less, ventral spines or sometimes (in varieties) with none; last dorsal plate without a median sulcus S. subspinipes Leach. SCOLOPENDRA GIGANTEA Linne. Syst. nat. ed. 10, 1758, 1, p. 638; Kraepelin, Revis. Scolop., 1903, p. 233; Brolemann, Cat. Myr. Bresil, 1909, p. 17. Scolopendra gigas Leach, Trans. Linn. soc. London, 1814, 11, p. 383; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 191. Scolopendra insignis Gervais et Goudot, Ann. Soc. ent. France, 1844, 2, p. 29. Scolopendra prasinipes Wood, Proc. Acad. nat. sci. Phil., 1862, p. 11. Scolopendra epilepiica Wood, Ibid. Scolopendra annulipes Lucas, Bull. Soc. ent. France, 1884, ser. 6, 4, p. 74; Brole- mann, Cat. Myr. Br6sil, 1909, p. 18. Localities. — State of x\mazonas: Obidos! James and Hunnewell, Nathaniel Thayer expedition. M. C. Z.; State of Para: Santarem! (Chas. Linden, M. C. Z.); State of Pernambuco: Villa Bella! (J. C. Fletcher; M. C. Z.). Lucas's type of S. annulipes was also from Brazil, the definite locality not being indicated. chamberlin: the ciiilopoda of brazil. 191 ScoLOPENDRA ANGULATA Newport. Ann. mag. nat. hist., 1844, 13, p. 97; Pocock, Journ. Linn. soc. London, 1893, 24, p. 146; Brolemann, Cat. Myr. Brdsil, 1909, p. 17. Scolopendra puncliceps Wood, Proc. Acad. nat. sci. Phil., 1862, p. 14. Scolopendra pundiscula Wood, Ibid. Scolopendra prasina C. L. Koch, Myr. 1864, 2, p. 23, fig. 146; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 192. Scolopendra nitida Porat, Bih. Svensk. vet. akad. Hand!., 1876, 4, no. 7, p. 8. Scolopendra respublicana Giebel, Zeitschr. gcs. naturw., 1879, 52, p. 326. Localities. — State of Matto Grosso : Madeira-Mamore, R. R. camp 41, on the Rio Madeira! (W. M. Mann); Brazil (without more definite locality, reported by Porat as nitida by Pocock and by Kraepelin) . The species is widely distributed elsewhere in South America and in the Antilles. Scolopendra viridicornis Newport. Ann. mag. nat. hist., 1844, 13, p. 97; KraepeUn, Revis. Scolop., 1903, p. 236; Brolemann, Cat. Myr. Bresil, 1909, p. 9. Scolopendra punciidens Newport, Loc. cit. Scolopendra cristata Newport, Loc. cit.; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 192. Scolopendra variegata Newport, Loc. cit., p. 98. Scolopendra hopei Gray, List Myr. Brit, mus., 1844, p. 45. Scolopendra herculeana C. L. Koch, Myr., 1863, 1, p. 22, fig. 20. Scolopeyidra morsitans C. L. Koch, Ibid., p. 37, fig. 33. IScolopendra costata C. L. Koch, Myr., 1863, 2, p. 25, fig. 147. Localities. — State of Rio Grande do Norte : Natal ! (W. M. Mann) ; State of Para: Para; State of Bahia : Bahia, Rio Sao Francisco (sec. Koch); State of Pernambuco ; State of Amazonas : Manaos, Amazon River! (J. C. Fletcher, M. C. Z.); State of Ceara: Ceara or For- taleza! (W.M.Mann); Brazil without definite locality (Chas. Linden, M. C. Z. ; and sec. Porat and Kohlrausch, as S. cristata; Gray, as S. hopei; and Newport and Kohlrausch, as viridicornis). Judging by the number of specimens secured by Mr. Mann, the species is very common at Natal. 192 bulletin: museum of comparative zoology. SCOLOPENDRA EXPLORANS, Sp. nOV. Dorsum dark olive, the plates being somewhat darker at the caudal borders and in a transverse row of areas or spots across the anterior half. Head darker excepting along the caudal border where it is much lighter, somewhat testaceous. Antennae olivaceous. Legs tes- taceous, excepting the posterior pairs, especially the ultimate, which are of a somewhat cherry -red color, particularly distad. Venter light brown. Prosternum dark brown; the prehensors somewhat rufous proximad of the black claws. Cephalic plate with two very fine submedian sulci diverging cepha- lad across entire length of plate. Very finely and subdensely uni- formly punctate. Antennae of moderate length. Composed of seventeen articles of which the first four are subglabrous and subdensely finely punctate; the other articles very densely clothed with fine brownish pubescence. Articles from the sixth distad not more than one and a half times longer than wide. Prosternum and prehensors subdensely punctate with fine punctae. With no median sulcus but with a distinct though fine transverse sulcus, caudad of which there is a mesal, shallow depression. Each dental plate with a large isolated tooth at ectal end and apparently three thoroughly fused teeth mesad of this, there being only slight indications of any divisions in the fused piece. First dorsal plate with a distinct cervical sulcus. Very finely punctate, the punctae being rather more scattered than on the head. No sulci detected on the second plate. The third to twentieth plates with paired longitudinal sulci extending full length of plate, fine. Finely punctate like the first plate, but the punctae becoming fainter and fainter caudad. First indications of lateral margination shown on the eleventh to fourteenth plate, the margination becoming more and more strongly marked caudad. Last dorsal plate without a median keel. Not punctate. Convexly arched on anterior portion, the posterior being more flattened. Caudal margin rather strongly bowed out caudad, the border being depressed in front of the median portion of the margin. Ventral plates either wholly without sulci or w4th very short traces of these at anterior border. Punctae fine, becoming faint caudad. Last ventral plate narrowed caudad. Caudal margin subtruncate, being but weakly rounded; mesally slightly notched or incurved. Without median longitudinal sulcus. chamberlin: the chilopoda of brazil. 193 Caudal process of coxopleurae of anal legs very short; ending in three or four spines or points, the coxopleurae being otherwise unarmed. Pores very fine and numerous. First pair of legs with two tarsal spines. Second to nineteenth pairs with a single tarsal spine. Prefemur of nineteenth legs dorsally at distal end with a single spine; that of the twentieth with two spines; the femur of the latter pair dorsally at distal end also with a single spine. Prefemur of twentieth legs unarmed ventrally. Prefemur of anal legs with the corner process at distal end above ending in two stout points or teeth; ventrally with five spines ar- ranged in two transverse or somewhat oblique rows, a distal row, composed of three spines, being at about the middle of length of joint, and a more proximal one of two spines (or in three longitudinal rows, 2, 2, 1); mesally with three or four spines in two longitudinal rows; and along dorsomesal edge with two spines in addition to a single one more strictly dorsal. Femur with two spines on proximal half along dorsomesal edge with a third one ventrad of these on the mesal surface; and, in addition, also a spine at distal end on mesodorsal corner. Claw with two basal spines which are very small. Length cir. 83 mm. Locality. — State of Matto Grosso: Madeira -Mamore R. R. camp 39, on the Rio Madeira! (W. M. Mann). This interesting species is evidently close to S. armata described by Kraepelin from Venezuela (Maracaibo), with which it is characteris- tically separated from all others now known. Among the more important dift'erences between these two species, so far as the descrip- tion of armata permits of comparison, may be mentioned the com- plete absence of any spines on the twentieth legs ; the dense pubescence of antennae on all articles distad of the fourth; the distinct punctation of the first dorsal plate; the margination of the dorsal plates from the eleventh to fourteenth caudad instead of from the eighteenth or nineteenth; the absence of paired sulci passing entirely across any of the plates; the presence of two spines at base of claw of anal legs instead of but one, etc. SCOLOPENDRA POLYMORPHA Wood. Proc. Acad. nat. sci. Phil., 1862, p. 11; Kraepelin, Revis. Scolop., 1893, p. 241. Scolopendra copeiana Wood, Journ. Acad. nat. sci. Phil., 1862, ser. 2, 6, p. 27; Pocock, Biol. Centr. Amer. Chilopoda, 1895, p. 19. Scolo-pendra mysteca Humbert et Saussure, Rev. mag. zool., 1869, ser. 2, 21, p. 157. 194 bulletin: museum of comparative zoology. Scolopendra pachypus KoWrausch, Archiv. naturg., 1881, 47, 1, p. 113. Scolopendra leptodera Kohlrausch Ibid.; Brolemann, Cat. Myr. Bresil, 1909, p. 19. Locality. — Brazil (type of leptodera Kohlrausch, which has been restudied by Kraepelin and identified with polymorpha) . This is a very common species in the southern United States and in Mexico. It has also been recorded from Venezuela. Scolopendra alternans Leach. Trans. Linn. soc. London, 1812, 11, p. 383; Meinert, Proc. Amer. philos. soc. 1886, 23, p. 193; Kraepelin, Revis. Scolop., 1903, p. 244; Brolemann, Cat. Myr. Bresil, 1909, p. 15. Scolopendra morsitans Palisot de Beauvois (non Linne), Ins. Afr. Amer., 1805- 1821, p. 152. Scolopendra sagraea Gervais, Ann. sci. nat., 1837, ser. 2, 7, p. 50; Brandt, Bull. sci. Acad, imper. sci. St. Peterburg, 1840, 7, p. 57. Scolopendra complanata Newport, Ann. mag. nat. hist., 1844, 13, p. 98. Scolopendra grayii Newport, Ibid. Scolopendra multispinata Newport, Ibid. Scolopendra incerta, Newport, Trans. Linn. soc. London, 1844, 19, p. 404. Scolopendra crudelis C. L. Koch, Syst. Myr., 1847, p. 387; Myr., 1864, 2, p. 36, fig. 158, 159; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 194. Scolopendra cubensis Saussure, Mem. Soc, phys. hist. nat. Geneve, 1860, 15, p. 387. Scolopendra testacea Wood, Journ. Acad. nat. sci. Phil., 1862, ser. 2, 5, p. 26. Scolopendra torquata Wood, Ibid. Scolopendra longipes Wood, Ibid. Scolopendra alternans Meinert, Proc Amer. philos. soc, 1886, 23, p. 193; Kraepelin, Revis. Scolop., 1903, p. 244; Brolemann, Cat. Myr. Bresil, 1909, p. 15. Locality.— Brazil ! (M. C. Z.). This is a very common species in the West Indies, etc. Scolopendra morsitans Linne. Syst. nat. ed. 10, 1758, 1, p. 638; Kraepelin, Revis. Scolop. 1903, p. 250; Brolemann, Cat. Myr. Bresil, 1909, p. 19. (For synonymy and bibliography cf. Kraepelin, Loc cit.). Localities. — State of Amazonas: Manaos; State of Parahyba: Parahyba! (Nathaniel Thayer expedition, M. C. Z.); State of chamberlin: the chilopoda of brazil. 195 Para: Para! (Nathaniel Thayer expedition, M. C. Z.); State of Bahia: Santarem! (M. C. Z.); State of Rio de Janeiro: Rio de Janeiro! (M. C. Z.); Brazil, without definite locality; (recorded by Gervais as brandtiana and plafypoidcs; by Saussure as brandtiana; by Porat as longicorms; also specimens in M. C. Z., with no more definite label). This is a cosmopolitan species. SCOLOPENDRA SUBSPINIPES Lcach. Trans. Linn. soc. London, 1814, 11, p. 3S3; Kraepelin, Revis. Scolop., 1903, p. 256; Brolemann, Cat. Myr. Bresil., 1909, p. 25. (For synonymy and bibliography cf. Kraepelin, Loc. cit.). Localities. — State of Bahia: Bahia! (Nathaniel Thayer expedition, M. C. Z.); State of Rio de Janeiro: Rio de Janeiro! (M. C. Z.); Brazil without definite locality (M. C. Z.); also recorded by Gervais as audax; by Newport as gervaisi, and placeae; by Koch, Kohlrausch, and Brolemann as armata). GEOPHILOIDEA. SCHENDYLIDAE. Of this family, the largest of the Geophiloidea in the known fauna, three genera are found in Brazil, Schendylurus, Adenoschendyla, and Thalthybius (Ballophilini). Schendylurus Silvestri. Mitth. Naturh. mus. Hamburg, 1907, 24, p. 245; Brolemann et Ribaut, Bull, soc. ent. France, 1911, p. 192; Arch. Mus. hist, nat., 1912, ser. 5, 4, p. 113. Schendyla Brolemann (ad. part, max.), Cat. Myr. Bresil, 1909, p. 6. The Brazilian species known may be separated by means of the key. Schendylurus brasilianus Silvestri, probably belonging to this genus, is not taken up, the published description being too meager to furnish sufficient information. 196 bulletin: museum of comparative zoology. Key to Species. a. Pairs of legs less than fifty. b. Ventral pores occurring on the fu-st sternite; pores on each sternite divided into three areas; cephalic plate scarcely longer than wide. Pairs of legs forty-one. S. ludcncahli Brolemann and Ribaut. bb. No ventral pores on the first sternite; ventral pores, at least on the anterior sternites, in an undivided area; cephalic plate considerably longer than wide. c. Antennae tliree times or more the length of the cephalic plate; last article of anal legs longer and conspicuously more slender than the penult; pairs of legs ( 9 ) 47. S. perditus, sp. nov. cc. Antennae but two times, or less, the length of the cephalic plate; last article of anal legs of nearly same length and thickness as the penult ; pairs of legs ( 9 ) thirty-seven. S. bakeri, sp. nov. aa. Pairs of legs near or above sixty. b. Prebasal plate not exposed S. gounellei Brolemann. bb. Prebasal plate exposed S. paulista Brolemann. ScHENDYLURUS LUDERWALDi Brolemann and Ribaut. Bull. Soc. ent. France, 1911, p. 220; Arch. Mus. hist, nat., 1912, ser. 5, 4, p. 117, fig. 48-52. Locality. — State of Rio de Janeiro : Campo Itatiaya (Mus. Paul, coll.). SCHENDYLURUS BAKERI, Sp. UOV. Very pale; more densely pigmented anteriorly, where the color is pale lemon-yellow. Head and the prosternum with prehensors chestnut or with slight tendency to ferruginous. Antennae similar to head but lighter. Body moderate; conspicuously narrowed from a little back of the middle caudad, but only moderately attenuated cephalad. Hairs of body and legs sparse, those of the legs chiefly toward the distal chamberlin: the chilopoda of brazil. 197 ends of articles as usual, these also being longer than the more proxi- mal ones. Cephalic plate widest in front of middle where it bulges convexly on each side; sides of head caudad of this straight and a little converg- ing to level of posterior end of first joint of prehensors (femur), then abruptly more strongly con^•erging to the caudal corners which are not rounded; caudal margin straight; anteriorly the head is convexly widely rounded. Longer than wide, the ratio being nearly 43:38. Antennae short being only 1.9— times longer than the cephalic plate; scarcely attenuated. Hairs very short, denser on the more distal articles, with hairs longer and more sparse on the proximal ones. Articles short, decreasing distad, with the sides more nearly straight than in perditus; ultimate article not much differing in length from the two preceding taken together. Prebasal plate exposed. Basal plate conspicuously narrowed cephalad; trapeziform. Two and a third times wider than long. Slightly more than one third as long as the cephalic plate (ratio 1:2.8-2.9). Claws of prehensorial feet when closed attaining the front margin of the cephalic plate. Joints all unarmed within as usual. Sides of prosternum for most of length nearly straight and but slightly converg- ing caudad, more abruptly rounding into caudal corners. Much wider than long, the ratio being 47:34. Longer than the first joint of pre- hensors in ratio 3:2. Dorsal plates mostly showing a fine median sulcus in addition to the lateral ones. Anterior prescuta short, those of the middle and pos- terior regions becoming rather long, the last few then again short. Spiracles all circular; the first considerably larger than the second, the others decreasing caudad and those of the posterior region very small or minute. First fourteen or fifteen sternites angularly produced at middle of caudal margin, the process small; process fitting into an excavation in the succeeding segment as usual. The anterior margin of the second sternite conspicuously extended from sides to middle, that of the third segment similarly but less strongly produced, that of the fourth merely convexly bowed out, and those of the succeeding ones straight, or nearly so, or even a little incurved. Ventral pores present on all sternites excepting the first and the last; pore area subcircular, with the pores numerous. Sternites mostly showing a longitudinal median furrow wliich is deepest just in front of the middle, and a weaker transverse furrow which curves across in front of the pore area. 198 bulletin: museum of comparative zoology. Last ventral plate very wide; trapezif orm ; the sides moderately converging caudad; caudal margin mesally incurved as in maiini (Plate 4, fig. 7). Coxopleural pores appearing as two large pits on each side, these being wholly covered by the last ventral plate excepting for a small ectal portion of each (Plate 4, fig. 7). First pair of legs a little more slender than the second, but not at all or but slightly shorter. Posterior pairs of legs longer and propor- tionately more slender than the anterior ones. Anal legs ( 9 ) much longer than the penult. Scarcely thickened. The ultimate article about equal in length to the preceding one and not more slender; but the last two articles together more slender than the tibia. Hairs more numerous than on other legs, especially on the proximal joints (Plate 4, fig. 7). Pairs of legs ( 9 ) forty -seven. Length 24 mm. Locality. — State of Amazonas: Manaos! (Mann and Baker). One female. Manifestly close in the main structural features to S. perditus. It is a materially larger form; has forty -seven pairs of legs as against thirty- seven in the latter species, has the antennae relatively much shorter; and the last article of the anal legs is proportionately much shorter and thicker as shown in the figures. SCHENDYLURUS PERDITUS, Sp. nOV. Body whitish, tinged with dilute lemon-yellow which is more evident anteriorly. Head and prosternum with prehensors ferrugi- nous. Antennae brownish yellow of faint ferruginous tinge. Moderate or slender; only slightly attenuated cephalad, more abruptly so caudad. Hair very sparse and mostly short over body, and those of legs also sparse. Cephalic plate evidently longer than wide, the ratio being about 5:4.4. Widest toward anterior end where the sides are convex; from this region caudad the sides are more nearly straight and con- verge at first moderately and then more abruptly toward the posterior corners; caudal margin appearing considerably incurved. Frontal suture not present. (Plate 4, fig. 1). Labral margin armed in the type with eighteen rather large, sub- acute, and strongly chitinous denticles, those at the sides being smaller, with apices turned mesad, and more acute than the more median ones. chamberlin: the chilopoda of brazil. 199 First maxillae of usual structure; outer branch robust, without lappets. Claw of palpus of second maxillae long, distally slender and acute and strongly curved; pectinate along both edges, the di\nsions long and slender. Antennae moderately long, being three and a fifth times as long as the cephalic plate, only a little attenuated distad. Articles of proxi- mal portion moderately long, each somewhat clavately widening from proximal and distad; the five articles preceding the ultimate short and relatively wider. Hairs of the last six or seven articles very short and rather numerous, on the more proximal articles becoming much more sparse and manifestly longer. Prebasal plate exposed. Basal plate trapeziform, the sides convex. About one third as long as the cephalic plate and 2.4-2.5 times wider than long. Claws of the prehensors when closed not fully attaining the front margin of the head. Claws smooth. All articles unarmed within as usual. Prosternum with sides for most of length nearly straight, a little converging caudad, more abruptly rounding mesad at posterior corners. Anterior margin well chitinized; but not at all denticulate; mesal incision between lateral portions shallow, semicircular; sloping from each side to the middle, there forming an obtuse reentrant angle. Prosternum much wider than long, the ratio being about 25 : 18; longer than the outer length of femur in about ratio 3:2. Hairs of pre- hensors sparse and in part moderately long; those of prosternum very sparse and short. (Plate 4, fig. 2). Prescuta of the anterior and of the posterior fourths of length short, the others being moderately long. Sulci sharply impressed. Spiracles all circular; the first conspicuously larger than the second, the others decreasing in size caudad and becoming very small or minute in the posterior region. The more anterior sternites with a rather narrow angular median caudal process which fits into a corresponding excavation in the succeeding sternite. Each with a subcruciform impression which is considerably expanded in the region where the furrows cross. Pores beginning on the second segment where there are from forty to forty- five in the type; pores present on all succeeding sternites excepting the last, those of the penult segment being fewer in number; pores arranged in an undivided circular area. Last ventral plate wide, trapeziform, the sides being nearly straight 200 bulletin: museltvi of comparative zoology. and converging caudad; caudal margin angularly excised at middle, convex laterally toward and about each caudal corner. (Plate 4, fig. 3). Each coxopleura with glands in the form of two large pits which are entirely simple and homogeneous; the anterior pore wholly and the posterior one mostly covered bv the last ventral plate. (Plate 4, fig. 3). First pair of legs a little more slender than the second but only slightly shorter. Posterior legs longer and proportionately more slender than the anterior ones. Anal legs much longer than the penult. Slender in the female. The distal joint somewhat longer that the preceding one and much more slender. Pairs of legs ( 9 ) thirty-seven. Locality. — State of Parahyba : Independencia ! liills north of the town. (Mann and Heath). ScHENDYLURUS GOUNELLEi (Brolemann). Schendyla gounellei Brolemann, Ann. Soc. ent. France, 1902, 71, p. 685; Cat. Myr. Bresil., 1909, p. 6. Schendylurus gounellei Brolemann et Ribaut, Arch. Mus. hist, nat., 1912, ser. 5, 4, p. 119, fig. 6, 62-67. Locality. — State of Sao Paulo: Fa^enda Nova Nicaragua. Schendylurus paulista (Brolemann). Schendyla paulista Brolemami, Rev. Museu Paulista, 1903, 6, p. 83, pi. 1, fig. 6-7; Cat. Myr. Bresil., 1909, p. 6. Locality. — State of Sao Paulo : Poco Grande. ^ In many ways close to the preceding species and possibly but a variety of it. (?) SCHENYDLURUS BILASILIANUS (SilvCStri). Nannophilus hrasilianus Silvestri, Ann. soc. ent. Belg., 1907, 41, p. 346. Schendyla brasiliana Brolemann, Cat. Mj'^r. Bresil., 1909, p. 6. Locality. — Brazil (precise locality not indicated). The generic position of this species cannot be determined from the chamberlin: the chilopoda of brazil. 201 original description. It appears not to be a true Nannophilus under which genus it was described, and is most probably a member of Schendvlurus. Adenoschendyla Brolemann and Ribaut. Bull. Soc. cnt. France, 1911, p. 192; Nouv. Arch. Mus. hist, nat., 1912, ser. 5, 4, p. 194. Three species and one variety of this genus, which is peculiar to tropical and subtropical America, are known from Brazil. Of these, one is here first described. The genus is close to Pectiniunguis. Pectiniunguis and Adeno- schendyla lack claws on the anal legs in contrast with species of the southwestern United States {montereus, heathi, etc.) The latter species differ as well in other respects and may be placed under a distinct genus to be known as Nyctunguis (P. montereus Chamb., type). Key to Species. a. Prebasal plate not exposed; none of the pore areas of the sternites are divided A. plusiodonta (Attems). aa. Prebasal plate exposed; some of the sternites of the median or posterior region of body longitudinally divided or geminate. b. Head much longer than wide (ratio 4:3 to 4:3.5), pairs of legs fifty-three to fifty-nine. c. Pores present on first sternite; head wider caudad than cephalad, longer than wide in ratio 4:3; pairs of legs 59 (9); length 40 mm A. parahybae, sp. nov. cc. No pores present on first sternite; head of same width anteriorly and posteriori}^, longer than wide in ratio 4: 3.5; pairs of legs 53 (cf )-55 ( 9 ) ; length 25 mm. A. geayi Brolemann and Ribaut. bb. Head but slightly longer than wide (ratio not more than 10:9), widest cephalad; pairs of legs forty -seven to fifty- three. c. Claw of palpus of second maxillae of usual form. A. imperfossa (Brolemann). cc. Claw of palpus of second maxillae globular at base. A. imperfossa holhonyx Brolemann and Ribaut. 202 bulletin: museltm of comparative zoology. Adenoschendyla parahybae, sp. nov. Mostly pale yellowish white, becoming more densely pigmented anteriorly, lemon-j-ellow. Head dilute ferruginous or orange, darker in a band immediately caudad of frontal region. Prosternum a little paler than the head, with the prehensors much lighter, yellow. An- tennae yellow. Rather slender with the bod}^ considerably attenuated cephalad and also very strongly at caudal end. Hairs sparse, of moderate length, more numerous caudad. Cephalic plate much longer than wide (4:3). Widest caudad, con- spicuously narrowed or constricted in frontal region at anterior end; anterior border subtriangular; caudal margin slightly concave; sides nearly straight from a little in front of caudal corners cephalad to frontal region. (Plate 5, fig. 1). Antennae strictly filiform as usual. Long, being a little more than three times the length of the cephalic plate. Articles mostly long, excepting those immediately preceding the ultimate. Ultimate article longer than the two preceding taken together. Hairs of articles of distal region very short, dense; those of proximal articles conspicu- ously longer and more sparse. Prebasal plate a little exposed, the cephalic plate not overlapping the basal. Basal plate with sides straight, strongly converging cephalad; three times wider than long. Claws of the prehensors when closed about even with the front margin of the cephalic plate; claws large and well overlapping; robust; articles all unarmed within as usual. Prosternum wider than long in about ratio 7:5; one and a half times longer than the outer height of the femur of prehensors; sides con- verging from the anterior corners to the caudal and straight excepting towards ends; anterior margin sloping a little caudad of directly mesad from the ectal ends to the mesal incision, which is shallow, laterally strongly chitinized but with no signs of teeth. Dorsal plates bisulcate; the sulci on the anterior plates distinct caudad as far as a fine transverse furrow a little in front of the caudal margin, this transverse furrow being in the form of a pair of concave impressions meeting at a cephalically directed angle; a fine median longitudinal sulcus also present as may also be one or two less distinct ones on each lateral part. Anterior prescuta short, the others in- creasing in length to the caudal region where they are moderately long, the last ones being again short. chamberlin: the chilopoda of brazil. 203 Spiracles all circular; the first one much larger than the second and the latter likewise considerably larger than the third; the others gradually decreasing caudad, in the posterior region becoming minute. The first sixteen sternites with the caudal border produced at the middle, the distinct process in each case fitting into a corresponding exca\ation in the succeeding segment, ^>ntral pores present on all sternites excepting the ultimate; on the sternite as far back as the twenty-fifth or twenty-sixth, the pores are in a single distinct sub- circular area; caudad of this the areas are more irregular, with a distinct tendency for each to become longitudinally divided into two areas or geminate. Last ventral plate very wide; sides nearly straight, strongly con- verging caudad; caudal corners rounded; caudal margin a little crenately incised a little each side of the middle. Rather densely clothed with fine short hairs, especially on the caudal portion. Coxopleurae subdensely clothed over the ventral area with fine short hairs. First pair of legs shorter and more slender than the second, the next few pairs gradually attaining the full size; anterior pairs of legs con- spicuously more robust than the posterior ones. Anal legs in the female much longer than the penult; very slender. Ultimate joint longer than the penult, very slender and ending in a minute membranous point but with no trace of a real claw. Hairs long and sparse. (Plate 5, fig. 3). Pairs of legs fifty-nine. Length 40 mm. Locality. — -State of Parahyba: Independencia ! (Mann and Heath). The present species differs from plusiodonta (Attems) in the much greater length and dift'erent shape of the cephalic plate, this in plusio- donta being only about as long as wide; in having the prebasal plate exposed; in the greater number of pairs of legs; in the character of the ventral pore areas, etc. The two species are similar in regard to the processes and pits of the anterior plates. Adenoschendyla plusiodonta (Attems). Pediniunguis plusiodontus Attems, Zool. jahrb. Syst., 1903, 18, p. 193, pi. 13, fig. 18; Chamberlin, Proc. Acad. nat. sci. Phil., 1904, p. 654. Adenoschendyla plusiodonta Brolemann et Ribaut, Nouv. Arch. Mas. hist, nat,, 1912, ser. 5, 4, p. 106. Locality. — State of Santa Catherina: Blumenau. 204 bulletin: museum of comparative zoology. Adenoschendyla imperfossa (Brolemann). Schendyla imperfossa Brolemann, Rev. Mus. Paulista, 1901, 5, p. 44, pi. 1, fig. 8-13; Cat. Myr. Bresil, 1909, p. 6. Adenoschendyla imperfossa Brolemann et Ribaut, Nouv. Arch. Mus. hist.nat. 1912, ser. 5, 4, p. 107. Locality. — Brazil. (Museii Paulista). Adenoschendyla imperfossa bolbonyx Brolemann and Ribaut. Brolemann et Ribaut, Nouv. Arch. Mus. hist, nat., 1912, ser. 5, 4, p. 107, fig. 18-23. Locality. — State of Sao Paulo (type Museu Paulista). Adenoschendyla geayi Brolemann and Ribaut. Bull. Soc. ent. France, 1911, p. 219; Nouv. arch. Mus. hist, nat., 1912, ser. 5, 4, p. 108, pi. 2, fig. 24-30, pi. 3, fig. 31-32. Locality. — State of Amazonas (Brazilian Guiana, Lower Carsevenne or Calyoene, Geay, collector). Thalthybius Attems. Zool. jahrb. Syst., 1900, 13, p. 139; 1903, 18, p. 183. Prionothalthybius Brolemann. Arch. zool. exp. et gen., 1909, ser. 5, 3, p. 334. Thalthybius (Prionothalthybius) perrieri Brolemann. Bull. Mus. hist, nat., 1909, p. 1, fig. 8-10. Locality. — State of Amazonas (Brazilian Guiana, Upper Carsevenne or Cal9oene; Geay, collector). CHAMBKKLIX: THE CIIILOPODA OF BRAZIL. 205 Of Uncertain Position. (?) Geophilus sublaevis Meinert. Natur. tiddskr., 1870, ser. 3, 7, p. 72. Locality. — State of ]\Iinas Geraes: Lagoa Santa. This is certain!}' not a true Geophilus, being in all probability a member of the present family. It seems likely to pro^•e to belong to Schendylurus. The anal legs are unarmed; the last ventral plate very wide with pores on coxopleurae said to be absent by Meinert but no doubt to be found after proper treatment with potash as has been shown to be true with -various species of Schendylurus, etc., which at first were considered to lack the pores; the presternum and joints of prehensors, unarmed, the claws of the latter not surpass- ing the front margin of the head; pairs of legs 67. Ortidae. Orphnaeus Meinert. Myr. Mus. Hauniensis, 1870, 1, p. 17; Proc. Amer. philos. soc, 1SS6, 23, p. 2.30; Zool. jahrb. Syst., 1903, 18, p. 200; Verhoeff. Bronn's Thierreich, 1908, 5, p. 294. Chomatohius Humbert et Saussure, Rev. mag. zool., 1870, p. 205; Miss, sclent. Mex., 1872, p. 145. Orphnaeus branneri, sp. nov. Dorsum yellowish brown, darker cephalad where the tergites are margined with a more deeply red stripe. The anterior dorsal plates, excepting the first one, with a conspicuous black spot on the anterior portion, this consisting typically of a narrow transverse stripe along the anterior margin connected at the middle by means of a broad neck with two short curved marks diverging from each other and bending out laterad near the middle of the plate; this mark in going from segment to segment caudad becoming less and less developed and finally disappearing entirely. No distinct geminate dark stripe such 206 bulletin: museum of comparative zoology. as is so characteristic of brevilabiatus. Basal and cephalic plates deep ferruginous; prosternum and prehensors ventrally similar but paler, the claws black. Antennae like cephalic plate but pale at their very tips. Venter pale testaceous, darker cephalad. Legs similar to venter. Body large and robust; strongly attenuated both cephalad and caudad. Hairs of body very fine and short as are also the few hairs of the legs. Cephalic plate conspicuously wider than long, the ratio being about 48:39. Widest caudad, where the sides are convex; moderately converging anteriorly in front of the middle to the anterior corners; lateral portions of anterior margin converging from the anterolateral corners to an obtuse angle at the middle; caudal margin widely, weakly con\'ex. Plate with subdense, uniform, fine punctae. Hairs very fine and short, numerous. First maxillae with inner division sharply set off; short and broad, apically rounded, not membranous. Outer division with the second and third articles not separated by a suture; short and thick; slightly membranous at tip on mesal side; membranous lappets of moderate length, the distal one wide and rather dorsal in position. Coxae of the second maxillae broadly joined at middle; with the usual oval opening toward the caudal end of each side. Claw of palpus rather small, bearing along each edge a fringe of about nine or ten spines. Antennae very short, being but 1.68 times the length of the cephalic plate. Flattened; very wide at base, then strongly narrowed, espe- cially distad of about the proximal fourth. Proximal articles very short, much wider than long, the more distal ones relatively longer; the ultimate article not much differing in length from that of the two preceding ones taken together. Hairs very fine and short, dense distad, becoming less so proximad. Prebasal plate not exposed. Basal plate embracing the cephalic. Very wide, with sides convex and not strongly converging cephalad ; nearly one half as long as the cephalic plate (ratio about 2.1 : 1), and very nearly three times as wide as long; finely and subdensely punctate like the head. Hairs similar to those of the head but considerably fewer in numl)er. Claws of the prehensors when closed very nearly but not wholly attaining the front margin of the cephalic plate; all joints unarmed; claws stout. Prosternum much wider than long, the ratio of width to length being 2.25-2.3:1; longer than the outer height of femur in ratio 25: 14; subdensely punctate as are also the proximal articles of the prehensors. (Plate 5, fig. 5). chamberlin: the ciiilopoda of brazil. 207 Dorsal plates with a pair of longitudinal deeply impressed sulci on the middle portion and an additional longitudinal sulcus farther laterad on each side less sharply impressed; indications on some segments also of a shallow median longitudinal median furrow. First dorsal plate considerably wider than the second and at its ends bent ventrad toward the base of each leg and crenately incised on the caudal side near each corner (Plate 5, fig. 4). ' Spiracles large; all elliptic, the anterior ones being oblique but more nearly horizontal than vertical, becoming strictly longitudinal caudad; first spiracle not larger than the second; those of the caudal region smaller as usual. Suprascutella large and distinct in the posterior region but absent in the anterior. Ventral pores in two broad transverse bands connected at the ends and thus forming a quadrangle. Enclosed area mostly with a distinct transverse furrow or row of impressed spots or the whole area rough- ened with irregular impressions, in most more deeply impressed longitudinally at middle. Last ventral plate pentagonal, the sides strongly converging caudad and the caudal margin straight. Marked with a longitudinal median furrow which is not especially deep. Coxopleurae rather large, coxiform; longer than thick in the direc- tion of thinnest diameter; the trochanter only about one third as long. Anal legs in male considerably shorter than the penult; composed of six articles ^ which decrease in diameter regularly from base distad. Pairs of legs 77. Length 88 mm.; length of antennae, 2.2 mm.; greatest width of body, 3.3 mm.; width of first dorsal plate L9 mm. Locality. — State of Rio Grande do Norte: Natal! (Mann). While this species is close in many features to brrvilabiatus, it is very easily separated from this wide-spread form. It is most readily dis- tinguished by the antennae which are much shorter, extending only to the caudal end of the basal plate or thereabouts, whereas in brevilabia- tus they reach upon or toward the caudal end of the second pediferous segment; also the antennae are conspicuously wider at the base and more strongly attenuated (Plate 5, fig. 4). The dorsal plate of the first segment is clearly different, being bent farther ventrad of ends and being more considerably notched on caudal side toward each caudolateral corner. It lacks the conspicuous geminate dorsal black stripe so characteristic of brevilabiatus. 1 The right leg of the type specimen appears to have been regenerated. It is shorter than the other and consists of but Ave articles. 208 bulletin: museum of comparative zoology. Orphnaeus brevilabiatus (Newport). Geophilus brevilabiatus Newport, Trans. Linn. soc. London, 1844, 19, p. 436. Geophilus lineatus Newport, Ibid. Geophilus guillemini Gervais, Insect. Apteres, 1847, 4, p. 311. Chomatobius brasilianus Humbert et Saussure, Rev. mag. zooL, 1870, p. 205; Miss, scient. Mex., 1872, p. 146, pi. 6, fig. 24. Orphnaeus brasiliensis Meinert, M}^. Mas. Hauniensis, 1870, 1, p. 20; Proc. Amer. philos. soc, 1886, 23, p. 232; Bollman, Proc. U. S. nat. mus., 1888, 11, p. 337; Brolemann, Mem. Soc. Zool. France, 1900, 13, p. 92; Ann. Soc. ent. France, 1902, 71, p. 652; Zool. anz., 1903, 26, p. 178; Rev. Mus. Paulista, 1903, 6, p. 71; Attems, Zool. jahrb. Syst., 1903, 18, p. 201; Cat. Myr. Brdsil, 1909, p. 5. Orphnaeus brevilabiatus Pocock, Journ. Linn. soc. London, 1893, 24, p. 472; Biol. Centr. Amer. Chilopoda, 1895, p. 40. Orphnaeus brasilianus nigropictus Attems, Loc. cit., p. 203. Localities. — State of Rio Grande do Norte: Ceara-Mirim! (Mann and Heath) ; State of Parahyba : Independencia ! (Mann and Heath) ; State of Amazonas: Manaos; State of Para: Para; State of Pernam- huco: Pernambuco; State of Rio de Janeiro: Rio de Janeiro!. Notiphilides Latzel. Myr. Ost-Ung. monarch., 1880, 1, p. 20; Zool. anz., 1880, 3, p. 546; Mein- ert, Proc. Amer. pliil. soc, 1886, 23, p. 233; Attems, Zool. jahrb. Syst., 1903, 18, p. 233; Verhoeff, Bronn's Thierreich, 1908, 5, p. 292. Notiphilides grandis Brolemann. Rev. Mus. Paulista, 1903, 6, p. 71, pi. 1, fig. 8-11. Locality. — State of Amazonas : Manaos. It was possibly a specimen of this species to which Cook gave the name Heniorya longissima; but as no description of the species is given both the generic and specific names, as Brolemann justly sug- gests, stand purely as nomina nuda. Aphilodontidae. Mecistauchenus Brolemann. Brasilophilus Brolemann, Bull. Soc. ent. France, 1907, p. 283. Verhoeff, Bronn's Thierreich, 1908, 5, p. 286. CHAMBERLIX: THE CIIILOPODA OF BRAZIL. 209 jMecistauchenus microxyx Brolcmann. Aphilodoii micronyx Brolemann, Rev. Mus. Paulista, 1901, 5, p. 46; Cat. Mjt. Br^sil., 1908, p. 3. Mecistauchenus micronyx Brolemann, Bull. ?^oc. ent. France, 1907, p. 283. Locality. — Brazil (precise locality not reported) . Aphilodon Silvestri. Comm. Mus. nac. Buenos Aires, 1898, 1, p. 39; Attem, Zool. jalirb. Syst., 1903, 18, p. 215, 283; Verhoeff, Bronn's Thierreich, 5, p. 279, 282; Silvestri, Boll. Lab. zool. R. sc. Agricol. Portici, 1909, 4, p. 53. Aphilodon angustatus Silvestri. Rend. R. accad. Lincei, ser. 5, 18, p. 269; Boll. Lab. zool. R. sc. Agricol. Portici, 1909, 4, p. 56. Locality. — State of IMatto Grosso: Urucum, Coruinba. Also reported from Paraguay and Argentina. Mecophilus Silvestri. Rend. R. accad. Lincei, 1909, ser. 5, 1, 18, p. 268. Mecophilus neotropicus Silvestri, Rend. R. accad. Lincei, 1909, ser. 5, 1, 18, p. 269. Locality. — State of Parana : Iguassti. Mecistoceph.\lidae. Mecistocephalus Newport. Proc. Zool. soc. London, 1842, p. 178; Trans. Linn. soc. London, 1844, 19, p. 276; Meinert, Naturli. tiddskr., 1870, ser. 3, 7, p. 92; Latzel, Myr. Ost-L^ng. monarch., 1880, 1, p. 160; Meinert, Proc. Amer, philos. soc, 1886, 23, p. 212; Haase, Abhandl. Mus. Dresden, 1887, 5, p. 100. Lamnonyx Attenis, Zool. jahrb. Syst., 1903, 18, p. 210; Verhoeff, Bronn's Theirreich, 1908, 5, p. 273. 210 BULLETIN": MUSEUM OF COMPARATIVE ZOuLOGY. Mecistocepil^lus puxctifrons Xew-port. Proc. Zool. soc. London, 1842, p. 179; Trans. Linn. soc. London, 1844, 19, p. 429; Meinert, Xaturh. tiddskr., 1870, ser. 3, 7, p. 97; Chamberlin, Ent. news, 1913, p. 122. Mecistocephalus guildingi Newport, Loc. cit.. p. 429; Meinert, Loc. cit., p. 96; Latzel, Mitt. Mus. Hamburg, 1895, 12, p. 101; Pocock, Trans. Linn. soc. London, 1893. 24. p. 470; Attems. Zool. jahrb. Syst., 1903, 18, p. 209. Mecistocephalus sulcicoUis Tomos^•ar^•, Termes. fiizetek, 1885, 5, p. 64. Lamnonyx punctifrons Attems, Loc. cit., p. 211. Locality. — State of Amazonas: IManaosI QIaiin and Baker). This appears to be the first record of the occurrence of a member of the Mecistocephahdae in Brazil. The species is common in the Bermudas and "West Inches. It must logicahy be regarded as the type of Mecistocephalus proper, the other species originally included imder tliis name by Xewport ha^*ing been removed to other genera. It would seem that Xe"n-port did not at the time he erected the genus, know or have in hand any species congeneric vnxh carniolcnsis and that in consequence there appears no justifiable way of continuing the prevalent practice of applying the name rvlecistocephalus to the genus including these species. As no difference of distinctive value has been pointed out between yunciifrons and guUdingii and as different authors refer to American specimens at times under one and at other times under the other name, I have united the two as one species as was long ago .suggested by ^Meinert. If the form occurring in the western hemisphere shall be found to differ definibly from that of the eastern hemisphere, it must bear the name guUdingii. Tygarrup, gen. nov. Body w-idest near middle, attenuated moderately cephalad and more strongly caudad. Head large. Cephalic plate longer than wide, narrowed caudad. Frontal stiture present. Antennae long, filiform. Cl\"peus proper large, triangularly extending forward in middle to between antennae, at middle being three times greatest length of labnmi; clothed with few hairs. Labrtmi tripartite, the median piece narrow and caudally one- chamberlin: the chilopoda of brazil. 211 toothed; the lateral pieces smooth, not longitudinally striate; not much bowed. From each anteroectal corner of labrum a suture extends obliquely cephaloectad, separating the median from the lateral divisions of the ventral portion of the cephalic plate; lateral division narrowed caudad, the mesal edge strongly chitinized and extended cephalad into an angular process as in related genera. ISIandibles with pectinate lamellae only. First maxillae with coxae completely separated, though closely appressed at median line. Inner branch clearly separated from coxa; subtriangular; distally prolonged into a conspicuous membranous lobe which is nearly as long as the proximal portion. Outer branch with second and third divisions completely coalesced; narrow; ex- tended distally into a long membranous lobe like that of inner divi- sion; no lappets present. Second maxillae with coxae rather short; coalesced at median line but more narrowly than in Mecistocephalus, etc. Pore of salivary gland on ectal portion of coxosternum near middle of length, not at caudal angle, the passage extending ectad to lateral margin. Palpi terminating in short, nearly straight, claws. Prehensors large, much exposed from above. Claws extending beyond front margin of head. Some of joints mesall}' armed. Basal plate narrow. Prebasal plate not exposed. Pleurae exposed at sides of basal plate. Dorsal plates bisulcate. Ventral pores absent. Last ventral plate subtriangular. Coxopleurae large, porose. Anal legs with six joints distad of coxopleurae; clawless. Anal pores present. Type. — T. intermedins, sp. nov. This genus is most closely related, apparently, to that embracing carniolensis, limatus, etc. (Mecistocephalus of most authors). It is different chiefly in the following points : — the materially greater shortness of the coxopleurae of the second maxillae and particularly the difference in position of the salivary pore, this being at about middle of length and toward lateral margin, not at extreme caudal angle as in Mecistocephalus, etc. ; and the lateral divisions of the lab- rum being unarmed, that is smooth, and not longitudinally striate. Also the hairs of labrum are much more sparse. From Mecistocepha- lus proper (piinctifrons Newport, etc.) it may readily be separated through the absence of the strongly chitinized process or tooth on the 212 bulletin: museum of comparative zoology. ventral border of head at anteroectal end of lateral division and the much shorter coxosternum of the second maxillae and difference in position of pore; the smaller middle piece of the labrum; and the larger clypeus, which extends cephalad in triangular form to near level of antennae. It agrees with the latter genus in the unarmed character of the lateral pieces of the labrum and also in the large size of the membranous lobe of the inner division of the first maxillae. Tygarrup intermedius, sp. nov. Yellowish; in type with a pale median longitudinal line paralleled or limited on each side by a somewhat darker stripe which is deepest cephalad and caudad, but these not evident in younger specimens. Head and prosternum with prehensors pale ferruginous. Antennae and legs pale. Head widest at level of labrum, conspicuously narrowed caudad with posterior corners well rounded; caudal margin straight; sides oblique and somewhat incurved from ends of frontal suture to ectal side of base of antennae; anterior margin substraight, narrowly semi- circularly excised at middle. Longer than wide in ratio 100:73. Pleural piece of ventral side of head plate with mesal edge strongly chitinized and ending cephalad in a pointed process as usual but with no trace of a tooth at cephaloectal angle. Clypeal region with hairs very sparse. Hairs of dorsal surface small and sparse. Labrum with the median piece very narrow, the sides of this being for most of length nearly parallel or but little converging caudad, its caudal end narrowing to a subacute tooth. Greatest length of labrum (i. e. at ends) about one third the median length of the clypeal area. Inner branch of the first maxillae with basal portion subtriangular in outline; the membranous distal division as long as or nearly as long as the basal, widening distad with mesal side concave and the ectal convex. Outer branch narrow, subcylindric, narrowing but moderately to the beginning of the long membranous distal division which is as long as the proximal division and widens distad like that of the inner division which it overtops by a short distance. Coxosternum of second maxillae rather short; median portion membranous; mesocaudal portion also membranous and not sharply defined. Pore close to outer edge, to which a passage from it leads, and a little caudad of middle of length, five or six bristles forming a row parallel with and a little removed from the anterior margin on each chamberlin: the chilopoda of brazil. 213 side. Claw of palpus small and pale, nearly straight, with a denticle on mesal side toward base. Antennae nearly 2.4 times longer than the head. Articles moderate in length, decreasing very gradually distad. Ultimate article (in mature specimen) shorter than the two preceding taken together in about the ratio 3: 4, in young specimens longer than these two. Claws of the prehensors when closed reaching to the end of the first antennal segment. Claw without a true tooth within, there being, however, a slight low, rounded, chitinous elevation; intermediate joints unarmed; prefemur (femuroid) at distal end on mesal side with a distally rounded tooth projecting cephalomesad. Prosternum a little wider cephalad than caudad; 1.3 times longer than wide; 1.9 times as long as length of prefemur on ectal side. Prosternum bearing on anterior margin each side of mesal incision a basally broad, conical tooth. Mesal incision with sides almost parallel, rounded at bottom. Basal plate a little overlapped both by head and by first dorsal plate; twice as wide as long; ratio of width at caudal end to that at anterior end as 45:34; head about 4.5 times longer. Prescuta of posterior and median region short; those of anterior region very short. Anterior ventral plates with a deep median longitudinal sulcus on caudal part and ending at about middle of plate in the angle of a short v or u-shaped impression, the arms of which diverge cephalad. This median sulcus becomes gradually weaker caudad, fading out and disappearing near the twenty-first segment, the u-shaped impression disappearing farther cephalad. Spiracles circular, rather large; first one largest, with the third considerably smaller and the second intermediate, the other decreas- ing gradually caudad as usual. First pair of legs greatly reduced, being only about two thirds the length of the second ones and much more slender. Anterior and posterior pairs not sensibly differing in length or thickness. Last ventral plate triangular or shield-shaped, the sides being convex; narrowly truncate at caudal apex. Coxopleurae moderately enlarged; each with two large pores partially covered by the ventral plate and over free ventral and lateral surface with regularly spaced, moderately numerous, smaller pores but these larger and fewer than those of Mecistocephalus, etc. Anal legs much longer than the penult but proportionately slender. Hair moderate in size, subsparse. Ultimate article narrowed distad, terminating in an obscure membranous tip. 214 bulletin: museum of comparative zoology. Length of type, cir. 18 mm. Locality. — British Guiana (taken at Washington, D. C, in pots of plants imported from that country). One adult, or nearly adult, and three adolescent specimens. Geophilidae. The new genus and species described below, and Ribaufia bouvieri Brolemann from the Carsevenne are the only representatives of this family, in the strict sense, at present known to occur within Brazil. Both genera belong to Chilenophilinae. Here also belongs Tai^iina, of which a representative from British Guiana is described. ScHizoNAMPA, gen. nov. Frontal suture not evident. Basal plate wide; overlapped by the cephalic plate. Antennae filiform. Dorsal plates bisulcate. Labrum free; tripartite; the median piece of moderate size, tri- angular, with the free edge armed with teeth; lateral pieces fringed with more slender, spinescent processes. Outer process of first maxillae uniarticulate ; bearing well-developed membranous lappets. Inner branch seemingly set off by suture; Coxae fused at middle. Second maxillae with coxae almost completely separated at mid- dle, the connection being narrow and membranous. Pleurosternal sutures strongly developed. Palpus triarticulate, terminating in a large simple claw. Femur bearing at distoectal corner a strongly chitinized acute process; tibia also bearing a similar process in a nearly corresponding, or slightly more dorsal position. (Plate 6, fig. 6). _ Prehensorial feet large, conspicuously exposed from above, and extending well beyond the front margin of the head; dentate within. Prosternum without distinctly developed chitinous lines. Basal plate trapeziform; wide. Ventral pores not evident. Last ventral plate wide. Coxopleural pores appearing as two large pits on each side. Anal pores not evident (in type). Anal legs with seven joints distad of the coxopleura, the small chamberlin: the chilopoda of brazil. 215 terminal article not bearing a claw. The extra article of the anal legs is at least strongly simulated on all the other legs, but especially the more posterior ones, by a contracted terminal division of the tarsus which for the most part is clearly distinct. (Plate 6, fig. 7). Genotype. — Schizojiamjya manni, sp. nov. This interesting genus is the second of the Chilenophilinae to be reported from Brazil, Ribautia being the first. Taiyuna, recorded from British Guiana, is the only other representative of the group at present known from South America. Schizonampa may readily be separated from the other known genera of the Chilenophilinae lacking a claw on the anal legs and having the small additional distal article as shown in the following key. Key to Genera of Section embracing Schizonampa. a. Ventral pores present; distomesal angle of coxa of second maxillae prolonged; first maxillae without lappets. Proschizotaenia Sil vestri . aa. Ventral pores absent; distomesal angle of coxa of second maxillae not prolonged; first maxillae with lappets. b. Pores occurring as two large pits on each coxopleura; femur and tibia of second maxillae prolonged into an acute, strongly chitinized process at distoectal angle. . Schizonampa, gen. nov. bb. Coxopleural pores small and isolated; tibia not prolonged at distoectal angle. c. First maxillae with two long membranous lappets on each side WaiophUus Chamberlin. cc. First maxillae with but a single lappet on each side, this being borne on the femur AUoschizotaenia Brolemann. Schizonampa manni, sp. nov. Slender; sides of bod}^ nearly parallel over most of length, but conspicuously attenuated at caudal end and moderately attenuated toward head. Body sparsely hirsute with short hairs; hairs of legs few, those present commonly arranged mostly toward the distal ends of articles. Color of body very pale, whitish yellow, the yellow being very dilute. Head with basal plate, prosternum, and prehensors, darker, somewhat light orange or dilute ferruginous; the antennae also similar. 216 bulletin: museum of comparative zoology. Cephalic plate much longer than wide (ratio about 4:29) narrowest cephalad; a little constricted in front of region where frontal suture would be if present, between which level and the caudal corners the sides are substraight or only very slightly convex; hairs sparse and mostly short or very short. Frontal plate not discrete. (Plate 6, fig. 1). Antennae short, being only 1.9 longer than the cephalic plate; attenuated. Articles mostly short, decreasing in size distad, those between the fifth and the ultimate being especially short; ultimate article longer than the two preceding taken together. Hairs on the first four or five articles moderate in length, sparse, those of the more distal articles considerably shorter and more dense. (Plate 6, fig. 1). Basal plate trapeziform as usual ; much overlapped by the cephalic plate as also by the first succeeding tergite. Exposed portion very short, at the median line being but one eighth as long as the cephalic plate and being about 4.5 or 4.6 times wider than long. (Plate 6, fig. 1,2). Clypeal region without any porose area; areolae distinct and uni- form excepting for a median area on the anterior portion in M'hich the areolae are conspicuously reduced in size and on which four hairs are borne, the clypeal region being elsewhere glabrous. Median piece of labrum rather large, triangular, bearing along the free margin five large acute and strongly chitinized teeth; lateral pieces with a fringe or more numerous slender spinescent processes. (Plate 6, fig. 4). First maxillae bearing ectally on each side one moderately long membranous lappet. Coxosternum mesally incised, but the coxae well fused for most of length of contact. (Plate 6, fig. 5). Coxae of second maxillae almost completely separated, there being only caudally a pale membranous connecting bridge. Pleurosternal sutures strongly developed. Coxa not at all produced at mesodistal angle. Femur and tibia bearing at distoectal angles a distinct, acute, well chitinized process, that of the tibia being somewhat more dorsal in position than that of the femur. Claw^ large and simple. (Plate 6, fig. 6). Claws of the prehensor when closed extending much beyond the anterior margin of the cephalic plate; attaining the distal end of the second antennal article. Claw not crenulate; armed at base with a stout tooth which is subtruncate distally. Intermediate articles without trace of teeth, but the femur bearing on mesal side toward chamberlin: the chilopoda of brazil. 217 distal end a stout, bluntl.^' rounded, tooth and also bulging in a small well-rounded eminence near the proximal third where there is indication of a suture such as is frecj^ucntly present (apparently of trochanter) . Lateral margin of prosternum parallel for most of their lengtli, in- curving only toward caudal ends. Prosternum a ver^- little wider than long (ratio cir. 18:17); longer than the femur on outer side in about ratio 17:11; anterior margin bearing two low and rounded, strongly chitinous, teeth; hairs very sparse. (Plate 6, fig. 3). The paired sulci of the tergites distinct; in addition to these a median longitudinal sulcus may be e\ident in the anterior region. Prescuta very short in the anterior region, becoming moderately long in the median and posterior regions. Hairs mostly very short. Spiracles all circular. The first one distinctly larger than the second, the others gi-adually decreasing caudad as usual. First pair of legs reduced, being shorter and decidedly more slender than the second. Posterior pairs of legs relatively but very little more slender than the anterior ones. A small third tarsal division simulat- ing or corresponding to the extra one of the anal legs is evident on all legs but especially the more posterior pairs; it is short and consider- ably more slender than the preceding one. Anterior ventral plates with a rather deeply impressed median longitudinal sulcus which extends entirely across the plate. First ten or eleven sternites produced caudad into a wide triangular process or lobe which fits into a recess in the anterior border of the succeeding plate in each case. Ventral pores not detected. Last ventral plate very wide ; trapeziform, the sides slightly convex anteriorly but substraight for most of length, strongly converging caudad; caudal margin straight. (Plate 6, fig. 7). ItiCoxopleural pores consisting of two large pits on each side; of these pits the anterior one each side is wholly covered by the ventral plate and the caudal one is covered excepting for a small portion. (Plate 6, fig. 7). Anal legs much longer than the penult; slender in the female. Second joint of tarsus long and slender; the third very short and abruptly more narrow. (Plate 6, fig. 7). Pairs of legs in the type thirty-seven. Length 13 mm. Locality. — State of Para : Para ! (Mann and Baker) . One female specimen was secured. 218 bulletin: museum of comparative zoology. RiBAUTiA Brolemann. Arch. zool. exp. et gen., 1909, ser. 5, 3, p. 335. RiBAUTiA BOUViERi Brolemann. Bull. Mus. hist, nat., 1909, p. 7, fig. 19-26. Locality. — State of Amazonas : Brazilian Guiana, on the upper Carsevenne or Cal^oene (Geay, collector). Taiyuna Chamberlin. Pomona college jour, ent., 1912, 4, p. 661. This genus was previously known only from California and Arizona in which states three species are known to occur. Taiyuna australis, sp. nov. Color yellow, of weak orange tinge cephalad. Head and prehensors darker, brown. Antennae similar to head, but paler distad. Body attenuated cephalad, more strongly so caudad; moderately robust. Head widest a little caudad of level of labrum from where the sides converge a little and are straight to the rounded posterior corners and also converge slightly cephalad to the similarly rounded anterior corners; posterior margin widely, somewhat flatly, convex; anterior margin with each side straight, extending from corner a little cephalad of mesad to the middle where there is a distinct notch. In type the head is 1.38+ times longer than wide. Inner branch of first maxilla terminating in an auriculiform mem- branous lobe at distoectal corner; bearing 3 to 5 bristles. Distal joint of outer branch long, apically rounded with the dorsoectal edge strongly chitinized; bearing about six bristles; membranous lappets long and spinulose, the distal one larger than the proximal, and ex- tending much beyond distal end of the outer branch. Coxae of second maxillae united at middle merely by a narrow mem- branous bridge; more strongly chitinized along edge below disto- mesal angle at which there is no trace of a process. Pleurosternal chamberlin: the chilopoda of brazil. 219 suture strongly marked as usual; salivary pore opening through the mesal border toward the anterior end in the usual way. Joints of palpus all without processes; femuroid narrowed proximally; claw small, simple. Antennae short, 1.8 times as long as head; attenuated. Articles decreasing uniformly in length from the first to the penult; ultimate article about equal in length to the two preceding ones taken together. Hairs of proximal articles sparse, distally becoming shorter and finer and more dense. Claws of prehensorial feet when closed reaching to or a little beyond distal end of first antennal joint; stout. Claw armed at base with a small, distally rounded tooth. Intermediate joints with slight, low and broad, chitinous denticles. First joint toward distal end and a little proximad of the corner with a thick rounded tooth. Anterior margin of prosternum unarmed; mesal incision very slight,^ semicircular. Prosternum wider than long in ratio 64:59; 1.64 times as long as outer length of prefemur; sides straight, only slightly con- verging from anterior end to the rounded caudal corners. Basal plate trapeziform, strongly narrowed cephalad; sides slightly convex caudad and slightly concave cephalad. 2.9 times as wide as median length, j as long as head. Overlapped both by cephalic and by first dorsal plate; the length of exposed portion to total length inclusive of covered ends as 3:4. Plate as a whole about 1.85 times wider than long. Anterior prescuta short, those of middle region becoming long and exceeding half the length of the major scuta, those of caudal region again short. Hairs short and sparse. Eight or ten of the first ventral plates with a triangular lobe on caudal border fitting into an excavation in the succeeding plate. Plates smooth, without pronounced furrows, or in some showing a shallow median longitudinal depression. Spiracles large, circular, or with the anterior ones very slightly longer dorsoventrally. First spiracle much exceeding the second in size, the others gradually decreasing caudad. Legs of first pair a little more slender than the second, only slightly shorter. Anterior and posterior pairs not at all or but little differing in length and robustness. Last ventral plate narrow; sides at first but slightly converging caudad, but more decidedly so toward posterior corners; caudal margin subtruncate. Depressed along the median longitudinal line. Coxopleurae moderately inflated; ventrally pierced by about 14 220 bulletin: museum of comparative zoology. or 16 pores, part of which are arranged along and partly covered by the last ventral plate; a pore somewhat larger than the others is isolated midway between the most caudal of the other pores and the distal end of coxopleura. Anal legs moderately long; slender, the joints decreasing in diame- ter from the femur distad; second tarsal joint moderately attenuated distad and bearing a very small and slender but distinct claw. Length about 42 mm. Locality. — British Guiana. (Taken at Washington, D. C, in pots of plants imported from that country). LITHOBIOMORPHA. LiTHOBIIDAE. LiTHOBius Leach. Trans. Linn. soc. London, 1814> 11, p. 381; Latzel, Myr. Ost-Ung. monarch., 1880, 1, p. 31; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 174; Ver- hoeff, Bronn's Thierreich, 1907, 5, p. 239. LiTHOBius FORFiCATUS (Linne). Scolopendra forficata Linne, Syst. nat. ed. 10, 1758, 1, p. 638. Liihobius forjicalus Stuxberg, Ofvers. Kong. vet. akad. Forh., 187.5, p. 27; Fedrizzi, Atti Soc. Ven-Trenk.. 1875, 5, p. 205; Latzel, Myr. Ost-Ung. monarch., 1880, 1, p. 57; Meinert, Proc. Amer. philos. soc, 1886, 23, p. 176. Lithobius parvolus Fedrizzi, Loc cit., p. 213. Lithobius trilineatus Brolemann, Cat. Myr. Bresil., 1908, p. 33. (For extended synonymy and bibliography cf. Stuxberg, Loc. cit.). Locality. — Recorded from Brazil as Lithobius trilineatus, the type being the only record of the species or genus from the country. It was probably introduced. SCUTIGEROMORPHA. SCUTIGERIDAE. Pselliophora Verhoeff. Sitz. Gesellsch. nat. freunde Berlin, 1904, p. 259; Bronn's Thierreich, 1907, 5, p. 230. chamberlin: the chilopoda of brazil. 221 PsELLiOPHORA NiGROViTTATA (Meinert). Scutigera nigroviltala Meinert, Proc. Amer. philos. soc, 1886, 23, p. 173; Po- cock, Biol. Centr. Amer., 1895, p. 650; Brolemann, Ann. Soc. ent. France, 1902, 71, p. 650; Brolemann, Cat. Myr. Br^sil, 1908, p. 34. Localities. — State of Rio Grande do Norte : Natal ! (W. M. Mann. Numerous specimens); State of Parahyba: Independencia ! (Mann and Heath); State of Matto Grosso: Madeira-Mamore R. R. camp 41, on the Rio Madeira! (W. M. Mann); State of Bahia: Santo An- tonio da Barra); Also Brazil, without special locality (M. C. Z.). Chamberlin. — Chilopoda of Brazil. PLATE 1. Cryptops heathi Chamberlin. 1 . Dorsal view of head and first dorsal plate. 2. Presternum and prehensors. 3. Anal leg. Mimops occidentalis, Chamberlin. 4. Dorsal view of head and first dorsal plate. 5. Prosternum and prehensorial feet. 6. Last dorsal plate. 7. Last ventral plate and coxopleurae. Hull. Mus. Coinp. Zool. Brazil. Chilopods Plate i HELIOTYPe CO., BObTCN. Chambeelin. — Chilopoda of Brazil. PLATE 2. Newportia longitarsis sylvae Chamberlin. 1. Dorsal view of head and first dorsal plates. 2. Anal legs with distal articles of tarsus omitted. Newportia paraensis, Chamberlin. 3. Anal leg. Newportia amazonica Brolemann. 4. Lateral view of distal end of tarsus showing claw. Newportia ernsti Pocock. 5. Lateral view of distal end of tarsus showing chitinous point or rudimentary claw occasionally present in some individuals. Otostigmus tidius Chamberlin. 6. Prosternum and prehensors. 7. Last dorsal plate. 8 1 Last ventral plate. Cupipes amazonae Chamberlin (of. Plate 3). 9 Last ventral plate and coxopleurae. Bull. Mus. Co in p. Zoo I. ]5nizil. Chilopods Plate 2 KELIOTYPe CO., BOSTON. Chamberlin. — CMopoda of Brazil. PLATE 3. Otostigmus amazonae Chamberlin. 1. Last ventral plate. 2. Last dorsal plate. Otostigmus suitus Chamberlin. 3. Last ventral plate. 4. Last dorsal plate. Cupipes amazonae Chamberlin (cf. Plate 2). 5. Prosternum (in part and prehensors). 6. Distal portion of anal leg showing claw, etc. Cupipes neglectus Chamberlin. 7. Prosternum (in part) and prehensors. 8. Last ventral plate and coxopleurae. Bull. Mus. Coinp. Zool. Brazil. Chilopods Plate 3 8 \ T /i 2 I HELIOTYPE CO,, BOSTON, Chambeblin. — Ohilopoda of Brazil. PLATE 4. Schendylurus perditus Chamberlin. 1. Cephalic region, dorsal view. 2. Cephalic region, ventral view. 3. Caudal region, ventral view. 4. Antenna (on same scale of magnification as fig. 1,2). Schendylurus bakeri Chamberlio. 5. Cephalic region, dorsal view. 6. Cephalic region, ventral view. 7. Caudal region, ventral view. 8. Antenna (on same scale of magnification as fig. 5, 6). Bull. Mus. Comp. Zool. Brazil. Chilopods Plate 4 ■^M- 1' -4- "^1^' 4/ / \ l:'l-i i' ir HELIOTYPE CO., BOSTON Chamberlin. — Chilopoda of Brazil. PLATE 5. Afienoschendyla parahybae Chamberlin. 1. Cephalic region, dorsal view. 2. Cephalic region, ventral view. 3. Caudal regions, ventral view. Orphnaeiis hranneri Chamberlin. 4. Cephalic region, dorsal view. 5. Cephalic region, ventral view. 6. Caudal region, ventral view. Bull. Mus. Comp. Zool. Brazil. Chilopods Plate 5 ^S^*-^'. HELIOTYPe CO., BOSTON. Chamberlin. — Chilopoda of Brazil. PLATE 6. Schizonampa mnnni Chamberlin. 1. Cephalic region, dorsal view. 2. Outline showing relations of basal plate. 3. Prehensors, ventral view. 4. Labrum. 5. First pair of maxillae. 6. Second rhaxillae. 7. Caudal region, ventral view. Bull. Mus. Comp. Zool. Brazil. Chilopods Plate 6 fe ■A 2 1 '. / ■. / k,. ---■^ k^. I't J? HELIOTVPE CO., BOSTON. The following Publications of the Museum of Comparative Zoology are in preparation: — LOUIS CABOT. Immature State of the Odonata, Part IV. E. L. MARK. Studies on Lepidosteus, continued. " On Arachnactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II. H. L. CLARK. The "Albatross" Hawaiian Echini. with 14 Plates. •Reports on the Results of Dredging Operations in 1877, 1878, 1879. and 1880. in charge of Alexander A GASsiz, by the U. S. Coast Survey Steamer "Blalie," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. Tlie Crustacea of the "Blake." A. E. VERRILL. The Alcyonaria of the "Blake." Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tanner, U. S. N.. Commanding, in charge of Alexander Agassiz, as follows: — K. BRANDT. The Sagittae. The Thalassicolae. O CARLGREN. The Actinarians. R. V. CHAMBERLIN. The Annelids. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes of Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. " The Scliizopods. HAROLD HEATH. Solenogaster. W. A. HERDMAN. The Ascidians. S. J. HICKSON. The Antipathids. E. L. MARK. Branchiocerianthus. JOHN MURRAY. The Bottom Speci- mens. P. SCHIEMENZ. Heteropods. THEO. STUDER The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. The Pteropods and The Alcyonarians. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Com- manding, as follows: — B. V. CHAMBERLIN. The Annelids. H. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. S. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. The Ophiurans. ■G. W. MULLER. The Ostracods. MARY J. RATHBUN. The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes. T. W. VAUGHAN. The Corals. Recent and Fossil. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. There have been published of the Bulletin Vols. I. to LIV. ; of the Memoirs, Vols. I. to XXIV., and also Vols. XXVL to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, XLL, and XLIV. Vols. LV. to LVIII. of the Bulletin, and Vols. XXV., XXX., XXXV., XXXIX., XL., XLIL, XLIIL, XLV. to XLVIII. of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation: — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S, N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical- Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905, Lieut. Com- mander L. M. Garrett, U. S. N., Commandbg. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director.. Contributions from the Geological Laboratory, Professor R. A. Daly, in charge- These publications are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology^ Cambridge, Mass. sv's'l Bulletin of the Museum of Comparative Zoblog:y AT HARVARD COLLEqt^. Vol. LVIII. No. 4. NOTES ON THE ONTOGENY OF PARADOX IDES, WITH THE DESCRIPTION OF A NEW SPECIES FROM BRAINTREE, MASS. By Percv E. Rai-moxd. With Oxe Plate. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. April. 1914. Reports on the iJciiiiNTiFic Results op the Expedition to the East- ern Tropical PAOTFrc, in charge of Alexander Agassiz, bt the U. S. Fish Commission Steamer "Albatross," from October, 1904, to March, 1905, Lieutenant Commander L. M. Garrett, U. S. N. Commanding, published or in preparation: — A. A. H. H. B. O. B. S. w L. W. c. 8. H. H. S w w c. AGASSIZ. V.' General Report on the Expedition. AGASSIZ. I.i Tliree Letters to Geo, M. Bowers, U. S. Fish Com. AGASSIZ and H. L. CLARK. The Echini. B. BIGELOW. XVI.n The Medusae. B. BIGELOW. XXIII.23 The Sipho- nophores. B. BIGELOW. XXVI.=« The Cteno- phores. P. BIGELOW. The Stomatopods. CARLGREN. The Actinaria. V. CHAMBERLIN. The Annelids. F.CLARKE. VIII. 8 The Hydrolds. . R. COE. The Nemerteans. J. COLE. XIX.i» The Pycnogonida. . H. DALL. XIV." The Mollusks. R. EASTMAN. VII.' The Sharks' Teeth. GARMAN. XII. 12 The Reptiles. J. HANSEN. The Cirripeds. J. HANSEN. XXVIl." The Schl- zopods. HENSHAW. The Insects. . E. HOYLE. The Cephalopods. C. KENDALL and L. RADCLIFFE. XXV. »B The Fishes. A. KOFOID. III.3 IX.» XX.'o The Protozoa. C. A. KOFOID and J. R. MICHENER. XXII." The Protozoa. C. A. KOFOID and E. J. RIGDEN XXIV." The Protozoa. P. KRUMBACH. The Sagittae. R.VONLENDENFELD. XXI.s< The Siliceous Sponges. The Holothurians. The Starfishes. — — The Opliiurans. G. W, MiJLLER. The Ostracods. JOHN MURRAY and G. V. LEE. XVI 1.1' The Bottom Specimens, MARYJ.RATHBUN. X.io The Crus- tacea Decapoda. HARRIET RICHARDSON. II.» The Isopods. W. E. RITTER. IV.« The Tunicates, ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. F. E. SCHULZE. XI.ii TheXenophyo- phoras. H. R. SIMROTH. The Pteropods and Heteropods. E. C. STARKS. XIII. » Atelaxia. TH. STUDER. The Alcyonaria. JH. THIELE. XV.I5 Bathysciadium. T. W. VAUGHAN. VI.« The Corals. R.WOLTERECK. XVIII." The Am- pliipods. » Bull. M. C. Z., Vol. XLVL. No. 4, April. 1905, 22 pp. 2 Bull. M. C. Z., VoL XLVL. No. 6, July, 1905, 4 pp.. 1 pi. ' Bull. M. C. Z . Vol. XLVL. No. 9, September, 1905, 5 pp.. 1 pi. « Bull. M. C. Z.. Vol. XLVL, No. 13, January. 1906, 22 pp., 3 pis. « Mem. M. C. Z., Vol. XXXIIL. January, 1906, 90 pp., 96 pis. « Bull. M. C. Z., Vol. L., No. 3, August, 1906, 14 pp., 10 pis. ' Bull. M. C. Z.. Vol. L., No. 4, November, 1906, 26 pp., 4 pis. « Mem. M. C. Z., Vol. XXXV., No. 1, February, 1907, 20 pp., 15 pis • Bull M. C. Z., Vol. L.. No. 6, February, 1907, 48 pp., 18 pis. •0 Mem. M. C. Z.. Vol. XXXV, No. 2. August, 1907, 56 pp., 9 pis. •I Bull. M. C. Z.. Vol. LI., No. 6. November, 1907, 22 pp., 1 pi. »J Bull. M. C. Z., Vol. LIL, No. 1. June, 1908, 14 pp., 1 pi. ts Bull. M. C. Z.. Vol. LIL, No. 2. July. 1908, 8 pp., 5 pis. >4 Bull. M. C. Z., Vol. XLIIL, No. 6, October, 1908, 285 pp., 22 pis. » Bull. M. C. Z., Vol. LIL. No. 5. October, 1908, 11 pp.. 2 pis. " Mem. M. C. Z.. Vol. XXXVII.. February, 1909, 243 pp., 48 pis. " Mem. M. C. Z.. Vol. XXXVIIL, No. 1, June, 1909, 172 pp., 5 pis., 3 maps. " Bull. M. C. Z., Vol. LIL, No. 9, June, 1909, 26 pp., 8 pis. «» BuU. M. C. Z., Vol. LIL. No. 11. August, 1909. 10 pp.. 3 pis. »Bull. M. C. Z., Vol. LIL, No. 13, September, 1909, 48 pp., 4 pis. « Mem. M. C. Z., Vol. XLI., August, September, 1910, 323 pp., 56 pis. 12 Bull. M. C. Z., Vol. LIV., No. 7. August, 1911, 38 pp. « Mem. M. C. Z.. Vol. XXXVIIL, No. 2, December, 1911, 232 pp., 32 pis. s* Bull. M. C. Z., Vol. LIV., No. 10. February, 1912, 16 pp., 2 pis. 25 Mem. M. C. Z., Vol. XXXV., No. 3, April, 1912, 98 pp., 8 pis. »> Bull. M. C. Z., Vol. LIV., No. 12, April, 1912, 38 pp., 2 pis. « Mem M.C. Z.. Vol. XXXV, No. 4. July. 1912, 124 pp.. 12 pis. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE. Vol. LVIII. No. 4. NOTES ON THE ONTOGENY OF PARADOX IDES, WITH THE DESCRIPTION OF A NEW SPECIES FROM BRAINTREE, MASS. By Percy E. Raymond. With One Plate. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. April, 1914. No. 4. — Notes oti the ontogeny of Paradoxides, with the description of a neio species from Braintree, Mass. By Percy E. Raymond. Paradoxides harlani is so large and striking a fossil, its occurrence is such an oasis in the sterility of Massachusetts palaeontology, and its discovery and subsequent history borders so closely upon the domain of romance, that it has become one of the most widely known of all the older invertebrate fossils. Although repeatedly described and figured antl known from abundant material, the species has never been studied from the phylogenetic standpoint, and it has not, there- fore, been brought out how strikingly different this species really is from most of the other members of the genus. In studying the speci- mens from the viewpoint of an investigation of the growth stages and relationship to other forms, the writer has been forced to the conclu- sion that there are really two species present in the Hayward quarry at Braintree, and the new species is now named in honor of the Messrs. Hayward, father and son, who have long been proprietors of the quarry which has furnished these trilobites, and who have served science by the care with which they have conserved good specimens. Paradoxides harlani Green. Plate, fig. 3-6. Paradoxides harlani Green, Amer. journ. soi., 1834, 25, p. 336. W. B. Rogers, Proc. Boston soc. nat. hist., 1856, 6, p. 27-29, 40-44. Stoddcr, Ibidem, p. 369. W. B. Rogers, Amer. journ. sci., 1856, ser. 2, 22, p. 296, Edinb. New philos. jour., 1856, new series, 4, p. 301 ; 1857, 6, p. 314. C. T. Ja'k- son, Compte.s rend. Acad. sci. Paris, 1856, 43, p. 883; 1858, 46, p. 254, Proc. Boston soc. nat. hist., 1859, 7, p. 54, 75. W. B. Rogers, Ibidem, p. 86. Ordway, Ibidem, 1860, 7, p. 427, 1861, 8, p. 1-5, fig. 2. C. T. Jackson, Ibidem, p. 58. Dana, Man. geol., 1863, p. 189, fig. 245. Walcott, Bulh 10, U. S. geol. surv., 1884, p. 45, pi. 7, fig. 3; pi. 8, fig. lb, c, e, (non 1, la, Id); pi. 9, fig. 1. Grabau, Occ. papers Bcston soc. nat. hist., 1900, 4, p. 681, pi. 35, fig. 3 (after Walcott); pi. 36 (the type); pi. 37; pi. 38, fig. lb, c, e; non 1, la, Id (after Walcott) ; pi. 39 (aft(>r Walcott). Shimer, Amer. journ. sci., 1907, ,ser. 4, 24, p. 178. Walcott, Smithsonian misc. coll., 1910, 53, p. 254, fig. 12, 13, (non 10, 11). 226 bulletin: museum of comparative zoology. Paradoxides spinosus W. B. Rogers, Geol. Penn., 1858, 2, p. 816, fig. 590. Barrande, Bull. Soc. geol. France, 1860, 17, p. 551; Proc. Boston soc. nat. hist., 1860, 7, p. 369. This species has often been described and is too well known to re- quire any formal description here; but I wish to emphasize certain features which, while now recognized, really have more importance than has previously been ascribed to them. As seen by the references cited above, Barrande considered P. harlam as identical with the Bohemian P. spinosus. This identification was immediately con- troverted by Ordway, and later writers have not accepted it; but of the two really vital differences of P. harlani from P. spinosus and most other species, only one has ever received attention. Ford (Amer. journ. sci., 1881, ser. 3, 22, p. 250) has called attention to the fact that the species of Paradoxides may be divided into two groups, in one of which the second segment of the thorax is always prolonged beyond the others, while in the other group the second segment is in no way distinguishable from the others. To the first group belong the Bo- hemian and South European species, while the Scandinavian, British, and American forms belong to the second group. Paradoxides spinosus has the second segment extended, while P. harlani has not. The second featin*e in which P. harlani differs from other species, and one which makes it almost unique, is the wide, depressed brim at the anterior end of the cranidium. Of the forty-six recognizable species of Paradoxides whose cranidium is known, only four, Para- doxides bennctti Salter, P. groomi Lapworth, P. regina INIatthew, and P. harlani Green have a rimless brim (though there is a possible fifth, P. hrachyrhachis Linnarsson). Of these, only two, P. harlani and P. regina have a wide brim in front of the glabella. All other species of Paradoxides described from adult specimens have the glabella reaching nearly or quite to the anterior margin. Among the numerous cranidia obtained from the Paradoxides beds at Braintree, there are some of the smaller ones which have a rim on the front of the cranidium, and the front of the glabella almost reaches the rim. These specimens have been considered by previous writers to be the young of P. harlani, and it was belie^'ed that in later stages of growth the anterior part of the cranidium became widened and flattened. Specimens recently obtained by the WTiter from Mr. Ha;y^vard's collection show that this could not have been the case, for there are specimens of the broad brimmed type which are of the same size or smaller than some of those showing the rim. The rimmed RAYMOND: NOTES ON THE ONTOGENY OF PARADOXIDES. 227 forms must therefore belong to another species, which is here de- scribed as P. haywardi. The hirgest cranidium of the rimmed form obtained is 35 mm. long, while the smallest cranidium with the brim and no rim is 19 mm. long (M. C. Z., No. 22, PI., fig. 3). On this specimen the part of the brim in front of the glabella is 2.25 mm. wide, or nearly 12'^"^ of the total length. On a specimen 84 mm. long it is 4.5 mm. long, or 1S%, while on a large cranidium, 103 nun. long, it is 12 mm. wide, or 11%. On the numerous cranidia between the smallest and largest it varies from 10% to 13% of the length, showing that while it grows wider during the growth of the individual, it is relatively about the same width in all cranidia above 19 mm. long. What it might be in smaller specimens we have as yet no means of knowing. The significance of this wide brim on the craniflium of P. harlani is best appreciated after studying the ontogeny of Paradoxides. Ontogeny of Paradoxides. The smallest specimen of Parado.xides known is that described by Barrande as Hydrocephalus satumoidcs (Systeme Silurien du centre de la Boheme, 1852, 1, p. 380, pi. 49). This specimen is slightly over 1 mm. long, the cephalon is oval, and makes up five sixths of the total length. The glabella is large, oval, makes up most of the cephalon, extends to the front of the head, and has no glabella furrows, though there is a median longitudinal furrow. The palpebral lobes form the lateral margins of the cephalon, but judging from the appearance of the cranidium, the free cheeks would have been present even at this early stage, had the specimens been complete. The occipital segment is prominently set off from the rest of the cephalon, and extends to the long, intergenal spines which cross it at right angles. One thoracic segment and a pygidium are present. F'rom the inferred presence of the free cheeks and the presence of a thoracic segment, it is evident that this is not a protaspis, but that several moults have already taken place. The second specimen described by Barrande is 1.33 mm. long, the Fig. 1. — Hydrocephalus saturnoides, show- ing two of the stages of development. After Barrande. Note the wide, oval glabella. Compare with Plate, flg. 9. 228 bulletin: museum of comparative zoology. cephalon occupying three fourths of the length and the thorax and pygidium one fourth. The thorax contains three segments. The cephalon has three glabellar furrows which cross the entire glabella. Three more stages of development are illustrated by Barrande, in the last of which the test is about 2 mm. long, the cephalon being a little over 1. mm. In the third, fourth, and fifth stages the first two seg- ments of the thorax bear backward-directed spines and the intergenal spines are still present. The most conspicuous feature of the cephalon in stage five is the presence of a narrow, smooth, flat brim on the front of the cranidium. This is first seen in stage four and becomes wider in stage five. The three glabellar furrows and the median longitudinal furrow are still present at stage five, but the median furrow is not so conspicuous as in the smaller specimens. The ontogeny of Olenellus indicates that the palpebral lobes are formed by the recurved pleura of the second glabellar lobes. It is very important to note that in these specimens known as Hydro- cephalus saturnoidcs the anterior ends of the palpebral lobes join the glabella in front of the anterior glabellar furrows, thus indicating that the furrows present are 2, 3, and 4. Beyond stage five Barrande did not trace any line, but the writer believes that Paradoxides orphanus and P. jynsiUus represent the next stages of this same species. In the M. C. Z. there is a cranidium 1.5 mm. long, identified as P. pusillus, but answering better to the description of P. orphcmus, which in some measure fills the gap between the largest of Barrande's figured speci- mens of H. saturnoides and the smallest of his P. orphanus and P. pusillus. In this specimen the anterior brim is narrow, occupying about the same proportion of the whole length as is shown in Bar- rande's figure of P. orphanus. Glabellar furrows 2, 3, and 4 all cross the whole glabella, as in //. saturfioides instead of 3 and 4 only, as in P. pusillus, but the connection between the two sides on furrow 2 is quite shallow. This specimen, moreover, adds another pair of fur- rows at the sides just in front of the palpebral lobes, as in P. pusillus. From this specimen to a typical pusillus with a wide brim, the collec- tion contains all stages, so there is no doubt of the connection in that direction. Barrande has figured {Loc. cit., 1872, 1, suppl. pi. 9, fig. 22, 23), an entire specimen of P. pusillus 2.5 mm. long, the cepha- lon of which makes up 55% of the length. Seven free segments are present, and the pygidium contains three or four more. The cephalon has a wide brim, 23% of the whole length, and there are no intergenal spines present, though the first two segments of the thorax have long spines, the spines of the second being longer than the first. Raymond: notes ox the ontogeny of paradoxides. 229 In the M. C. Z. there is a complete specimen of P. " pimllus " which is 4.5 mm. long, or about twice as large as the one figured by Barrande. The cephalon is 2 mm. long, or 44% of the whole length, and the brim is narrower than in the last specimen. Both the genal spines and those of the second thoracic segment are long, but the first thoracic segment has lost its spines. There are about fifteen thoracic segments ending in free spines, but those back of the tenth are crowded into an extremely small space. (See Plate, fig. 9). The largest cranidium of P. jmsillus in the collection is 4.5 mm. long, and Barrande does not mention any larger. In this largest specimen the brim is only .5 mm. wide, thus occupying but 13% of the length, showing that with increase in size the brim is becoming shortened again. Furrows 3 and 4 cross the glabella, while 2 does not. Fur- row 1 is present and distinct at the sides. Next to this specimen stands our smallest cranidium of P. rugidosus Hawle and Corda, which is 4 mm. long and practically identical with the largest of P. lyusiUus, but furrows 2 are a little more faint, furrows 4 turn more obliquely backward, and the posterior ends of the palpe- bral lobes are a little closer to the glabella. From this small specimen we have all gradations up to a full-grown P. rugulosus with a crani- dium 27 mm. long. In the adult P. nirjulosus the anterior furrow is very narrow, the glabella being almost in contact with the rim. Whether this line of development is based entirely upon one species or not, the fact remains that in the development of the brim of Para- doxides there is a change from the very youngest where there is na brim to a youthful stage where the brim is wide, then back to a later adult stage in which the brim is again diminished almost to nothing. In the matter of the brim, therefore, P. harlani retains at maturity a youthful characteristic, lost in P. rugulosus when less than 10 mm. long. There is a certain amount of e^^dence that the line traced above from Hydrocephalus saturnoidcs through Paradoxides orphanus and P. pusillu^ to P. rugidosus represents the growth stages of one species. There are in the collections at the M. C. Z. specimens in all stages be- tween P. pitsillus and the adult P. rugulosus, and the only sharp break is between the largest specimen of Hydrocephalus saturnoidcs and the smallest of P. orphanus or P. pusillus. In the matter of the brim there is no break, for we see it gradually becoming wider and wider in speci- mens of H. saturnoides, it continues getting wider in P. orphanus and P. pusillus up to a certain stage, then decreases in width in the larger pusillus and the young of rugulosus. The only great change between 230 bulletin: museum of comparative zoology. Hydrocephalus saturnoides and P. orphanus is the introduction just at this point of the anterior pair of glabellar furrows and the reduction in size of the glabella. But the furrows come in where we would have predicted them, just in front of the palpebral lobes. And this brings out a point, which was wholly unexpected, that in this species of Paradoxides the glabella of the youngest specimens known is smooth and unsegmented, and gains its furrows during growth. This will be referred to again later. There are also external indications which indicated that the above series may be a natural one. Hydrocejyhalus saturnoides, P. orphanus, and P. pusillus all palpably represent immature individuals and have been generally so considered. Hydrocephalus was placed by Barrande as akin to Paradoxides, and by Beecher was referred to that genus. Barrande separated it from Paradoxides, first, on account of the course of the facial suture, which left the genal spine on the fixed instead of the free cheeks ; second, on account of the longitudinal furrow on the glabella; and third, because of the few thoracic segments, there never being more than twelve. The third characteristic merely indicates immature specimens and need not be considered. In regard to the second one of the remarka- ble features of P harlani is that this same longitudinal furrow of the glabella is present, at least as a line of weakness, in very large speci- mens. As to the first, the spines on the fixed cheeks are merely the terminal spines of the palpebral lobes, the intergenal or " interocular " spines known also in the young of Olenellus. From the form of the cranidium it is evident that entire specimens had free cheeks, and they doubtless bore the true genal spines. There is, therefore, no reason for separating Hydrocephalus from Paradoxides. In regard to occurrence, all four species come from the Cambrian band in the vicinity of Skrey, and so far as known are found together in the same beds. Most of the specimens of H. saturnoides in the M. C. Z. collection are from Teirovic, from which locality there are also specimens of P. pusillus and P. rugulosus. All are in the same kind of matrix and have the same sort of preservation. At Slap, where P. rugulosus is most abundant, H. saturnoides seems less com- mon, though P. pusillus is quite common. Of the Bohemian species, P. sacheri Barrande, P. lyelli Barrande, P. bohemicus (Boeck) , and P. rotundatus Barrande seem to be confined to the Ginetz band of the Cambrian, hence it is unlikely that these young should belong to any of those species. Paradoxides spinosus (Boeck) is very common in the band of Skrey and when I began to RAYMOND : NOTES ON THE ONTOGENY OF PARADOXIDES. 231 study these specimens, I supposed them to be the young of that species. That they do not belong to that species is, however, shown convincingly by the eyes. The adult of P. rugulosus has the palpebral lobes touching the glabella at their anterior ends and reaching the occipital furrow behind, while in the adult of P. spinosus the eye is much shorter, and does not reach either the glabella or the occipital furrow. Barrande has figured {Loc. cit., pi. 12, fig. 7) a specimen of P. spinosus with a glabella 4.5 mm. long in which the eyes have the same position as in the adult, while in the Museum series, specimens this size have the rugulosus type of long eyes. The thorax of P. pu-sillu-s is of the sj/inosus rather than the rugulusus type, but that is a character which might change readily during growth from a size of 4 mm. up to the size of the adult P. rugulosus. It is, however, possible that the young of the two species would be, in the earliest stages, in- distinguishable. Young Specimens of Paradoxides. Bohemia. Hydrocephalus carens Barrande and Paradoxides inflatus Corda make a short series showing the early growth stages of some as yet unidenti- fied species. Barrande figures nine stages in the growth of //. carens, the smallest specimen being 2 mm. long and the largest 4 mm. The thorax and pygidium together show three segments in the smallest Fig. 2. Fig. 3. Fig. 2. — Hydrocephalus carens'BQ.TTdLXide. After Barrande. This series shows two stages in the development. Compare with Plato, fig. 3. Fig. 3. — - Paradoxides inflatus Corda. After Barrande. and fifteen in the largest. The glabella is almost circular and shows no glabellar furrows in the first four stages described, and only No. 4 is present in the last five. Intergenal spines are present on all, and on specimens 6-9 the first two segments of the thorax have terminal 232 bulletin: museum of comparative zoology. spines, those of the first segment being the longer. Barrande figures an entire specimen of P. inflaius Corda, 5 mm. long, which differs from //. carens only in possessing free cheeks, glabellar furrow No. 3, and in having the terminal spines of the first thoracic segment reduced to normal, while the second pair have increased in length. Into what species this form finally developed there is no way of determining with- out more material. It is interesting to note, however, that the devel- opment agrees with the series described above in that the glabella is first smooth, and the glabellar furroAvs are added during the nepionic stages. They seem to be greatly retarded in this form, as only two furrows (Nos. 3 and 4) have been formed in the largest specimen fig- ured by Barrande. It may be noted that in this form, as in Hydro- cephalus saiurnoides, the brim on the cranidium widens constantly during the known stage of growth, though it never achieves any great width. The M. C. Z., contains a single specimen of P. inflaius (No. 651) about 5 mm. long from Velka in the Cambrian band of Ginetz. All the other specimens recorded have come from the band of Skrey, but that it does occur in the more southern band suggests that it may possibly be the young of some form which in the adult has only two pairs of glabellar furrows, possibly P. bohemicus (Boeck). The M. C. Z. contains a single minute specimen (No. 33) 1 mm. long of a young "Hydrocephalus" which is in many respects quite unlike H. saturnoides. (Plate, fig. 8). The specimen difi^ers from that form in having the glabella narrow^, expanding forward. The occipital ring and furrow are well marked, the glabellar furrows 2, 3, and 4 are deeply impressed, extend across the glabella and divide it into ring- like lobes. The anterior lobe, which is composed of lobes 1 and 2, is transversely oval, and nearly twice as wide as that portion of the glabella back of it. Furrows 1 are faintly indicated, and the frontal lobe has a deep longitudinal furrow which does not reach lobe 3. The palpebral lobes are long, and extend into spines behind the occipital segment. The brim is narrow, and the truncation at the sides is so slight that it is doubtful if free cheeks were present. This specimen is about the size of the smallest specimen of H. saturnoides, and it does not seem that it could represent a younger stage in the development of that species, but it appears probable that it is the protaspis of another form. From the outline of the posterior end of the ispecimen, it seems probable that it is complete, and that the small posterior pro- jection represents the proto-pygidium. The young specimen of P. spinosus figured by Barrande has already been alluded to. Its chief interest lies in the fact that so small a raymoxd: notes on the ontockav of paradoxides. 233 specimen, only 1 1 mm. long, should be so like the adult. The chief differences are that the eyes are slightly longer, the genal spines arise further forward on the head, and the terminal spines of the second thoracic segment are much longer. These terminal spines of the second segment seem the most persistent of the juvenile characters, and as has already been stated, all the Bohemian species have some remnant of these spines in the adult stage. The youngest specimen of P. hohemicus yet seen is that figured by Barrande {Loc. cit., pi. 10, fig. 25). It is 14 mm. long, not including spines, and exhibits only two youthful characteristics. The eyes are long, and the terminal spines of the second thoracic segment are greatly prolonged. The glabella shows only two furrows (Nos. 3 and 4) the same as in the adult. Outside Bohemia young Paradoxides are evidently exceedingly rare and have been figured only incidentally. Nothing except a few cranidia seems to have been found. Scandinavia. The "youngest Scandinavian specimen known is that figured by Linnarsson as the t^'pe of his species P. aculeatus. This is a cranidium slightlv less than 2 mm. long from the Paradoxides oelandicus zone of Borgholm. There is a relatively wide brim in front of the glabella, the palpebral lobes extend from the second glabellar lobes to the pos- terior margin, intergenal spines are present, the glabella is long and narrow, expands slightly forward and has four pairs of glabellar fur- rows. It resembles a young Paradoxides more than Hydrocephalus saturnoides does, as the glabella is of a more normal shape. This specimen ^ is probably the young of either P. sjogreni or P. oelandicus, more likely of the former, as Linnarsson figures another small crani- dium 5.5 mm. long which he refers to that species, though with doubt. This latter specimen also has a wide brim and four pairs of furrows. The adults of both P. sjogreni and P. oelandicus have a narrow border, and the glabella almost touches the rim. Great Britain. Salter figures (Quart, journ. geol. soc, 1869, 25, pi. 3, fig. 8-10) three young specimens of Paradoxides hicksi. These specimens are I Lindstroem considers this the young of P. oelandicus. K. Svensk. vet.-akad. Handl., 1901, 34, no. 8, p. 17. 234 bulletin: museum of comparative zoology. young cranidia in the pusillus stage. They have four pairs of glabellar furrows, long eyes, and a wide brim. The adult has four pairs of furrows, short eyes well forward, and no brim. Hicks, in the descrip- tion, says of these specimens : — "In the young the margin is equal all around, and a considerable space, also, separates the glabella from the anterior margin. This space gradually diminishes as the individual grows; and the glabella enlarges until, as in the fully grown species, the margin becomes fully obliterated." America. Paradoxides tenellus Billings from Newfoundland is another form described from a young specimen in the pusillus stage. The typical cranidium is 6 mm. long, shows four pairs of furrows on the glabella, has long palpebral lobes and a wide brim. This species may possibly be the young of P. decorus, a very imperfectly known species which Billings describes from a cranidium about 26 mm. long, and which has four pairs of furrows, but has the glabella in contact with the rim. It occurs in the same locality as P. tenellus. Matthew presents notes on the young of P. cteminicus and P. acadius, and figures three specimens of these species in the pusillus stage. All have wide brims, long palpebral lobes, and four pairs of furrows. They are the young of forms whose glabellas nearly or quite touch the anterior rim, and which retain the four pairs of furrows and long palpebral lobes at maturity. Summary on Ontogeny. From the above survey of the material now available for the study of the ontogeny of Paradoxides, we see that the youngest shell or protaspis is very similar to that of Olenellus. The glabella in the youngest specimens of both species of "Hydrocephalus" is specialized and unlike that of any other trilobite of which the young is known, in that it occupies a large part of the head, is very wide, and bears no transverse furrows. The first furrow to appear is a median longi- tudinal one, which is obliterated at an early stage. Glabellar furrows are introduced in young stages, and in later stages of development there seems to be no reduction of furrows by their obliteration suc- cessively from the front backward, such as is seen in some of the later trilobites. The glabella occupies the whole length of the cranidium r-^ymoxd: notes ox the ontogeny of paradoxides. 235 in the youngest stages knoAvn, becomes proportionately shorter (hiring some of the early nepionic stages (pusillvs stages), and becomes longer again in the neanic and early ephebic stages. The palpebral lobes are in general very much longer in young stages than in later ones, but many species are primitive in this regard, and retain the long eyes at maturity (P. rugulosius group). Most of the adult characteristics are assumed at an early age, so that specimens 6-10 mm. long are often almost identical in form with the adult; but certain minor features such as the lateral extention of the second thoracic segment, persist well on into the ephebic stages. ApjjUcaiion to P. harlani. It will now be seen why the form of the brim of this species is so important. The wide brim is a feature Avhich, in this genus, is decidedly larval in character, and in such forms as are knowTi to have had it, it is lost at an early age, when the cepha- lon was 6-10 mm. long. Its retention in large adults like P. harlani is most unusual. Another result arrived at above is applicable to P. harlani. It was found that the glabellar furrows were not lost by the adult, but that, on the contrary, the adult had more furrows than the young. None of the very young of P. harlani are known, but the smallest glabellas now before us (11 mm. long) show two pairs of furrows which cross the glabella and another pair, (Xo. 2), which are faintly indicated at the sides. The small cranidium figured (Plate, fig. 3) which is 19 mm. long, shows a similar condition, but the No. 2 furrows are much more distinct. In some of the largest specimens (glabella 100 mm. long) furrows 1 and 2 are both distinct, and most specimens with cranidia more than 40 mm. long show all four pairs of furrows. In these two features, then, the wide brim and the slow ac- quisition of glabellar furrows this species is very primitive. The palpebral lobes in the smallest specimen mentioned above reach from the glabella back to the occipital furrow, and their chord is 6 mm. in length. In specimen No. 22 they meet the glabella, but terminate 1 mm. in front of the occipital furrow, and the chord of the lobe is 7 mm. In the adult this eye is proportionately much smaller, for, on a cranidium 79 mm. long the posterior end of the lobe is 9 mm. from the occipital furrow and 8 mm. from the glabella, the chord of the lobe being 21 mm. Thus the proportion of the length of the chord of the palpebral lobe to the length of the cranidium in the smallest specimen is .50, in the second, .32, and in the adult, .26, or a reduction of about one half. In common with most other species of Paradoxides, P. harlani shows a great lateral extension of the fixed cheeks during the process of growth. 236 bulletin: museum of comparative zoology. Eight of the hirgest cranidia in the collection, varying from 60 to 120 mm. long, show a longitudinal cracking along the median line which strongly recalls the median longitudinal furrow of Hydro- cephalus. In this case it is not exactly a furrow, but the crushing along this line of so many specimens indicates a line of weakness here. The backward or forward turning of the third and fourth furrows of the glabella at the median line in so many species is also probably to be connected with this furrow. Comparison with other Species. As mentioned above, there are only a few rimless species of Para- doxides, and it is with such forms alone that P. harlani can be compared. It has been compared most commonly with P. spinosus (Boeck) ; but from that species it differs, not only in the possession of a rimless brim and the absence of the terminal spines on the second thoracic segment, but also in the pygidium, which in P. harlani is larger and longer and has a much longer axial lobe than the Bohemian species. The species which lack the rim, besides P. harlani, are P. bennetti Salter, P. groomi Lapworth, and P. regina Matthew. P. bennetti is very similar to P. harlani in the shape of the glabella, the possession of four pairs of glabellar furrows, and medium sized eyes. The genal spines appear to be shorter, and according to the single specimen in the M. C. Z., the brim is not so wide. In this specimen, the second segment does not seem to be enlarged as indicated by Salter and mentioned by Ford, but is actually smaller than the first. Paradoxides groomi is known only from fragments which indicate a species similar to P. harlani, but with narrower thorax and, accord- ing to Cobbold's description, shorter fixed cheeks. The principal differences between P. regina Matthew and P. harlani seems to lie in the pygidium, which is more quadrangular in outline and has a shorter axial lobe in the former species than in the latter. Outside the pygidium it is, as has been pointed out by Grabau, ex- ceedingly difficult to point out differences between the two species. The majority of specimens of P. harlani have a narrower cephalon and glabella than the Acadian form, but as Grabau has already shown in his table of measurements, we have specimens of a wide form which correspond very closely to the dimensions of Dr. Matthew's specimen. Incidentally I might mention that the Geological section of the M. C. Z. has recently acquired a specimen of P. harlani with a cranidium 138 RAYMOND: NOTES ON THE ONTOGENY OF PARADOXIDES. 237 mm. long, or IS iiiin. longer than the cranidiuuT of Dr. ^latthew's specimen of P. regina. This, restored on the basis of the dimensions of the wide form, would exceed both in width and length the P. regina; but, unfortunately it appears to belong to the narrow type, and would therfore cover considerably less area than that species. Paradoxides haywardi sp. nov. Plate, fig. 1, 2, 7. Paradoxides harlani Walcott, partim, Bull. 10, U. S. geol. surv., 1884, pi. 8, figs, 1, la. Id. Among the many cranidia which have been collected at the Hay- ward quarry at Braintree there are a number which at the anterior end differ markedly from Paradoxides hnrlani. Instead of having a flattened rimless brim, they have an ele\ated striated marginal rim separated from the glabella by a narrow furrow. Moreover, the out- line of the margin of the anterior end of the cranidium, instead of being a smooth curve as in P. harlani, is obtusely pointed, the two segments of the rim being straight and meeting at an angle of about 150° in front of the axial line of the glabella. The glabella is convex, semicircular in front, widest opposite the anterior ends of the palpe- bral lobes. The dorsal furrows are strongly marked and come to- gether in front of the glabella. No specimen so far seen shows any traces of glabellar furrows No. 1, and No. 2 when present, are only slightly impressed and do not show at all in most specimens. No. 3 are quite strong but usually do not meet at the centre, though in one ^or two specimens they appear to. No. 4 extend across the glabella. The occipital ring is wide and bears a small median tubercle. The eyes are of medium size for the genus. The palpebral lobe does not reach the occipital furrow behind, nor is it connected with the glabella at the front. No free cheek has been seen which can be assigned with certainty to this species. Free cheeks seem to be less common than any other parts of trilobites at the Braintree locality. One specimen in the collection shows a part of a thorax still connected with a cranidium, though the body is partially shoved under the head. Eleven segments are present. The axial lobe is about one third the total width, and the pleura are marked by wide grooves which cross them diagonally, the grooves extending out to 238 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. the point where the pleura begin to taper into spines. These grooves occupy a much larger proportion of each pleuron than do those of P. harlani. The terminal spines appear to be rather longer and more slender than in P. harlani. The second spine is not longer than the others. The pygidium differs strikingly from that of P. harlani, and is the one figured by Walcott {Loc. cit., pi. 8, fig. Id). This pygidium differs from the one found on entire specimens of P. harlani in being broader than long, and in having the axial lobe distinctly rounded in- stead of triangular. The last five segments of this specimen (M. C. Z. No. 20) are much longer than those on the average specimen of P. harlani, and it is seen from the small entire specimen in the collection of Mr. W. P. Haynes that it belongs to P. haywardi. A second pygidium of this type is in the M. C. Z. (No. 652) and a third in the Geological section of the M. C. Z. A nearly entire specimen which is in the collection of Mr. Winthrop P. Ha;\Ties has seventeen thoracic segments and shows that the short pygidium assigned above to this species really belongs to it. Mr. Haynes's specimen is largely exfoli- ated, and the substance of the pygidium is entirely gone. Its outline is, however, indicated on the matrix, and it is of the broad short type, apparently 9 mm. long and 15 mm. broad. The whole trilobite is about 105 mm. long and is quite narrow^ the thorax being about 55 mm. wide at the first segment. Types. — As the holotj'pe of this species I have selected cranidium No. 16 (Plate, fig. 1) recently presented to theM. C. Z. by Mr. Lemuel Hayward. The left side of this cranidium is very well preserved, but the right side is broken and the broken piece partially thrust under the glabella. Furrow 3 is distorted. As paratypes I have selected specimens Nos. 17, and 18, both o% which are illustrated. All are in the M. C. Z. Measuremejits. Holotype. — Cranidium 35 mm. long; glabella 31 mm. long, 23 mm. wide at front of eyes; chord of palpebral lobe 13 mm. long; rim 2 mm. wide at front of glabella, 4 mm. wide at corner of cranidium. Paratypes. — No. 17. Cranidium 32 mm. long; glabella 30 mm. long, 22 mm. wide at front of eyes. No. 18. Cranidium and eleven thoracic segments, 76 mm. long as it stands, but as some segments are pushed under the head, the actual length was probably about 80 mm.; width of axial lobe at 2d segment 20 mm.; total width of thorax, not including terminal spines, 58 mm. RAYMOND: NOTES ON THE ONTOGENY OF PAIi.\DOXIDES. 239 Remarks. As may be seen by the following quotation, the specimens here separated as a new species have not escaped observation. Dr. Wal- cott (Loc. cit., p. 46), in discussing the broad and narrow forms of P. harlani says : — " 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 smallest 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." It is not the narrow form of P. harlani as described by Walcott and Grabau which I am separating as a new species, but the form with the narrow brim and raised, striated rim. Judging from the above quotation, this form has been placed as the young of the narrow form of /'. harlani. As has been shown under the description of P. harlani above, material recently collected shows that the young of P. harlani had a broad flat rimless brim, similar to that of the adult, so that the rimmed forms can not be referred to that species. Comparison with other Species. Paradoxides hay war di is a much more normal type of Paradoxides than P. harlani, and it is therefore comparable to a far greater number of species. From P. harlani itself, it differs, as has already been pointed out, in having an angular instead of a rounded frontal margin, and in having a narrow brim and thickened rim on the front of the cranidium, in the absence of the anterior pair of glabellar furrows, and probably in the wider furrows and narrower spines on the pleura of the thorax. It resembles P. etcminicus Matthew more closely than any other American form, but differs from that species in having shorter eyes, the lobes of which do not touch the glabella or neck ring, in lacking the anterior pair of glabellar furrows, and in having a wider groove separating the glabella from the rim. Most of these same dif- ferences and others obtain between P. abenacus Matthew, P. acadius Matthew, P. micrnac Hartt, P. lamellatus Hartt, and P. haywardi. P. regina Matthew and P. hennetti Salter appear to have the wide margin of P. harlani; and of Billings's two Newfoundland species, P. tenellus seems to be based on immature specimens, and P. dccorus is not well known. 240 bulletin: museum of comparative zoology. Of the numerous British species, P. aurora Salter, P. hicksi Salter, and P. forchhammeri Angelin have the glabella reaching on to the rim, P. davidis is of the P. tessini type with the elongated terminal spines of the pleura, as is also, presumably, P. bohemicus salopiensis Cobbold, of which the thorax is not loiown. P. groomi Lapworth is of the P. harlani type, P. harknessi Hicks agrees with the new species in the presence of a furrow between the glabella and rim, but the eye lobes are much longer and more narrow, and the glabella is narrower and retains the first pair of furrows. P. intermcdius Cobbold is quite similar in form of glabella, groove and rim, to P. haywardi, but the palpebral lobes are too long and reach both the neck ring and the glabella. The Paradoxides rugulosus Hawle and Corda, Cobbold, also has the long eye lobes and the first pair of glabellar furrows, and, moreover, differs from the true rugulosus in lacking the furrow which should separate the glabella from the rim. Turning now to the Scandinavian species, we find that P. forch- hammeri Angelin, P. hicksi palpebrosus Linnarsson, P. oelandicus Sjogren, and P. tumidus Angelin all have the glabella reaching the rim, and most of them have other features in which they differ strikingly from P. haywardi, while P. affinis Angelin, and P. tessini Brongniart and its varieties of course have the long terminal spines on the pleura. P. tuberculatus Angelin is known only from a fragment which has a large tubercle on the fixed cheek opposite the basal lobes of the glabella. P. brachyrhachis Linnarsson appears to be a rimless species with four pairs of glabellar furrows, and so comparable to P. harlani, while P. aculeatus is based on a very immature specimen. There is, therefore, no Scandinavian species very closely allied to P. haywardi. Of the Bohemian species, P. bohemicus (Boeck) is quickly eliminated because in the adult the glabella reaches the rim and the terminal segments of the pleura are elongated. P. desiderahis is probably not a Paradoxides, but possibly an Albertella, and P. ciycctans also is doubt- fully a Paradoxides. P. imperialis is known only from a fragment of the thorax, while P. inflatus Corda, P. pusillus Barrande, and P. orphanus Barrande are evidently based on very immature specimens. P. spinosus and P. rotundatus Barrande both have glabellas which in the adult touch the anterior rim and retain all four pairs of furrows in most cases. P. rugulosus and P. sacheri both have the groove in front of the glabella, but P. rugulosus has very long eye lobes, touching the glabella and occipital ring, while P. sacheri has very short diagonal furrows and very curving spines on the thorax. P. lyelli has a long narrow glabella which touches the marginal rim. ILA.YMOXD: NOTES ON THE ONTOGENY OF PARADOXIDES. 241 Only a few forms are known from southern Europe (Spain, France, Sardinia). P. asper Borneniann is founded on fragments and its right to be called a Paradoxides is queried by Pompeckj. P. mcdiferraneus Pompeckj is very similar to P. rugulosiis, — was so identified by Ber- geron,— and the cephalon is therefore similar to that of P. haywardi. P. barrandei Barrois has the whole four pairs of glabellar furrows and the glabella touches the marginal rim, but P. prodoanu^ de Verneuil and Barrande, which is very similar, has a narrow furrow between the glabella and rim, but the eyes are very close to the glabella, their anterior ends touch it, and the posterior ends also curve in unusually close to the glabella. It appears then that P. haywardi is most closely allied to P. etemini- cus Matthew of the St. John area in New Brunswick, P. Intermedius Cobbold from Comley in Stropshire, England, and P. rugidosus Hawle and Corda, and P. mediterraneus Pompeckj of central and southern Europe. These four species, so far as they are known, all seem to be- long to the P. rugidosus group in which the eye lobes are very long, the glabella is separated from the marginal rim by a furrow (P. ete- minicus has a very narrow furrow) and have a rather long pygidium, the posterior margin of which is straight or concave in outline (the pygidium of P. intermedius is an exception). The eyes of P. haywardi are not of the P. rugulosiis type, nor is the short wide pygidium. It may be noted, however, that the pygidium is not very different from that of P. intermedius Cobbold, to which P. haywardi seems on the whole to be most closely allied. 242 bulletin: museltm of comparative zoology. Bibliography. The following list contains a reference to the first or, in the case of a few species, a subsequent easily accessible description of the species of Paradoxides referred to in this article. Massachv^etts. Paradoxides harlani Green, Amer. journ. sci., 1834, 25, p. 336. P. haywardi RajTnond, ante, p. 237. New Brunsivick. P. abenicus Matthew, Trans. Roj'. soc. Canada, 1885, 3, sect. 4, p. 78, pi. 7, fig. 17a-d. P. acadius Matthew, Ibidem, 1882, 1, sect. 4, p. 103, pi. 9, fig. 16-18. P. acadius suricus Matthew, Ibidem, 1885, 3, sect. 4, p. 77, pi. 7, fig. 16. P. eteminicus Matthew, Ibidem, 1882, 1, sect. 4, p. 92, pi. 9, fig. 7-12. P. eteminicus brevialus Matthew, Ibidem, p. 99, pi. 9, fig. 1-3. P. eteminicus malicitus Matthew, Ibidem, p. 101, pi. 9, fig. 13, 13a. P. eteminicus pontificalis Matthew, Ibidem, p. 102, pi. 9, fig. 15, 15a; 1890, 8, sect. 4, p. 136, pi. 11, fig. 8. P. eteminicus quacoensis Matthew, Ibidem, p. 101, pi. 9, fig. 14, 14a. P. eteminicus suricoides Matthew, Ibidem, p. 97, pi. 9, fig. 4-6. P. lamellalus Hartt, Dawson's Acadian geology, 1868, 2d ed., p. 656. Walcott, BuU. 10, U. S. geol. surv., 1884, p. 25, pi. 3, fig. 2-2a. P. lamellatus loricatus Matthew, Trans. Roy. soc. Canada, 1882, 1, sect. 4, p. 106, pi. 9, fig. 19. P. micmac Hartt, Dawson's Acadian geology, 1868, 2d ed., fig. on p. 657. No description. Matthew, Trans. Roy. soc. Can., 1885, 3, p. 80, pi. 7, fig. 18. P. micmac pontificalis Matthew = P. eteminicus pontificalis . P. regina Matthew, Trans. Roy. soc. Canada, 1887, 5, sect. 4, p. 119, pi. 3. Newfoundland. Paradoxides bennetti Salter, Quart, journ. Geol. soc. London, 1859, 15, p. 552, text fig. 1. P. decorus Bilhngs, Paleozoic fossils of Canada, 1874, 2, pt. 1, p. 75. P. tenellus Bilhngs, Ibidem, p. 74, fig. 43. ratmond: notes on the ontogeny of paradoxides. 243 Wales and England. Paradoxides aurora Salter, Quart, journ. Geol. soc. London, 1869, 25, p. 54, pi. 2, fig. 9-12. P. hohemicus salopiensis Cobbold, Ibidem, 1913, 69, p. 45, pi. 4, fig. 6-17. P. davidis Salter, Ibidem, 1S64, 19, p. 275, fig. Cobbold, Ibidem, 1911, 67, p. 285, pi. 24, fig. 17a, 17b, 18. P. forchhammeri Angelin. Salter Mem. Geol. surv. United Kingdom. Figures & descriptions Brit, organic remains, 1864, dec. 11, pi. 10, fig. 9. P. grootni Lapworth, Geol. mag., 1891, dec. 3, 7, p. 532, footnote. Cobbold, Quart, journ. Geol. soc. London, 1911, 67, p. 283, pi. 23. P. harknessi Hicks, Ibidem, 1871, 27, p. 399, pi. 15, fig. 9-11. P. hicksi Salter, Ibidem, 1869, 25, p. 55, pi. 3, fig. 1-10. Cobbold, Ibidem, 1913, 69, p. 47, pi. 4, fig. 1-5. P. interinediits Cobbold, Ibidem, p. 29, pi. 2, fig. la-lc, 3, ? fig. 2, 4-1 Ic. P. rugidosiis Hawle and Corda, Cobbold, Ibidem, 1911, 67, p. 286, pi. 24, fig. 14-16C. P. spp. ind. Cobbold, Ibidem, p. 285, pi. 24, fig. l-7c. Scandinaiia. Paradoxides aculeatus Linnarsson, Sveriges geologiska undersokning, 1877, ser. c, no. 22, p. 8, pi. 1, fig. 11. (See p. 233). P. affiins AngeUn, Palaeontologia Scandinavica, 1878, 3d ed., appendix, p. 94, pi. la, figs. 3, 3a. P. brarhyrhachis Linnarsson, Loc. cit., 1884, ser. c, no. 54, p. 16, pi. 3, fig. 6-10. P. forchhammeri Angelin, Palaeontologia Scandinavica, 1852, pars 1, p. 2, pi. 2, fig. 1-3. P. hicksi palpebrosu^ Linnarsson, Sveriges geologiska undersokning, 1879, ser. c, no. 35, p. 9, pi. 1, fig. 5-11. P. oelandicus Sjogren, Geol. foreningens Stockholm Forhandl. 1872, 1, no. 5, p. 72, pi. 5, fig. 1. P. sjogreni Linnarsson, Sveriges geologiska undersokning, 1877, ser. c, no. 22, p. 6. pi. 1, fig. 7-9 and ? 10. P. tessini Brongniart, Hist. nat. crustaces fossiles, 1822, p. 31, pi. 4, fig. 1. Angelin, Palaeontologia Scandinavica, 1852, pars 1, p. 1, pi. 1, fig. 1. P. tessini oelandicus Sjogren, Angelin, Loc. cit., 1878, 3d ed., appendix, p. 94, pi. la, fig. 2, 2a-2c. P. tessini wahlenbergii Angelin, Ibidem, p. 94, pi. la, fig. 1, la, lb. P. tuberculatus Angelin, Ibidem, p. 94, pi. la, fig. 4. P. tumidus Angelin, Ibidem, p. 95, pi. 3, fig. 2, 2a. 244 bulletin: museum of comparative zoology. Bohemia. Paradoxides hohemicus (Boeck), Mag. for naturv., 1828, 8, heft 1, pi. 2, fig. 11. Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 367, pi. 10, 22-25. P. expectans Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 918, pi. 13, fig. 10. P. imperialis Barrande, Ibidem, 1852, 1, p. 373, pi. 13, fig. 19. P. inflatus Hawle and Corda, Prodrom einer Monographic, etc. 1847, p. 32. P. lyeUi Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 917, pi. 8, fig. 1. P. orphanus Barrande, Ibidem, p. 373, pi. 13, fig. 11-13. P. pusilhis Barrande, Ibidem, p. 374, pi. 13, fig. 14, 15. P. rolundatus Barrande, Ibidem, p. 371, pi. 14, fig. 24. P. rugulosus Hawle and Corda, Prodrom einer monographic, etc. 1847, p. 32. P. sacheri Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 369, pi. 9, fig. 30. P. spinosus (Boeck), Mag. for naturv., 1828, 8, heft 1, pi. 2, fig. 12. Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 370, pi. 11, fig. 1; pi. 12, fig. 1-14; pi. 13, fig. 1, 2. Hydrocephalus carens Barrande, Systeme Silurien du centre de la Boheme, 1852, 1, p. 377, pi. 49, figs. H. saturnoides Barrande, Ibidem, p. 380, pi. 49, figs. Southern Europe. Paradoxides asper Bornemann, Nova acta K. Leop. Carol, d. akad. naturf., 1891, 56, p. 468, pi. 39, fig. 1, la, ?2, ?4. Not a Paradoxides according to Pompeckj. P. harrandei Barrois, Mem. Soc. geol. Nord. Lille, 1882, 2, p. 169, pi. 4, fig. la-lf. P. mediterraneus Pompeckj, Zeitschr. Deutschen geol. geseUsch., 1901, 53, heft 1, p. 2, pi. 1, fig. 1-3. P. pradoanus Verneuil and Barrande, Bull. Geol. soc. France, 1860, ser. 2, 17, p. 526, pi. 6, fig. 1-6. Raymond. — Paradoxides. EXPLANATION OF PLATE. Fig. 1. Paradoxides haywardi Raymond. The holotype. Hayward quarry, Braintree, Mass. Nat. size. No. 16, M. C. Z. Fig. 2. The same species. A cranidium from the same locality. Nat. size. No. 17, M.C. Z. Fig. 3. Paradoxides harlani Green. The cranidium of a young specimen from the same locality as the above. Nat. size. No. 22, M. C. Z. Fig. 4. The same species. An imperfect cranidium of about the same size as the cranidia of P. haywardi shown in fig. 1, 2. Same locality. Nat. size. No. 23, M. C. Z. Fig. 5, 6. Paradoxides harlani ? Green. * The free cheeks of young individuals, referred to P. harlani rather than to P. haywardi, merely on account of the greater abundance of the first named species. Same locaUty as above. Nat. size. No. 24, 25, M. C. Z. Fig. 7. Paradoxides haywardi Raj'mond. A specimen retaining a part of the thorax. The cranidium is imperfect and has been pushed back over the thorax. A part of an overturned free cheek is present at the side. Same locality. Nat. size. No. 18, M. C. Z. Fig. 8. Paradoxides sp. ind. A protaspis from the Middle Cambrian at Teirovic, Bohemia. X 16. No. 33, M. C. Z., Schary coll. (The median depression is too long in the figure). Fig. 9. Paradoxides rugulosus Hawle and Corda. A young individual from Slap, near Skrey, Bohemia. X 8. No. 423, M. C. Z., Schary coll. Bull. Mus. Cotnp. Zoiil Paracioxides r . /■ , ^ V-j ".ill . U HEIIOTYPE CO., BOSTON The following Publications of the Museum of Comparative Zoology are in preparation: — LOUIS CABOT. Immature State of the Odonata, Part IV. E. L. MARK. Studies on Lepidosteus, continued. On Arachnactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II.. with 14 Plates. H. L. CLARK. The "Albatross" Hawaiian Echini. Reports on the Results of Dredging Operations ii? 1877, 1S78. 1879. and 1880. in charge of ALEX.4NDER Agassiz, bj' tlic U. S. Coast Survey Steamer "Blalie," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. The Crustacea of the A. E. VERRILL. The Alcyonaria of the "Blake." Blake. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tamner, U. S. N., Commanding, in charge of Alexander Agassi z, as follows: — K. BRANDT. The Sagittae. The Thalassicolae. O CARLGREN. The Actinarians. R. V. CHAMBERLIN. The Annelids. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes of Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. " The Schlzopods. HAROLD HEATH. Solenogaster. W. A. HERDMAN. The Ascidians. S. J. HICKSON. The .Yntipathids. E. L. MARK. Brauchiocerlantluis. JOHN MURRAY. The Bottom Speci- mens. P. SCHIEMENZ. The Pteropods and Heteropods. THEO. STUDER. The Alcyonarians. The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. Reports on the Scientiflc Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Com- manding, as follows: — R. V. CHAMBERLIN. The Annelids. H. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. 8. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. The Ophiurans. <5. W. MtJLLER. The Ostracods. MARY J. RATHBUN. The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes. T. W. VAUGHAN. The Corals, Recent and Fossil. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. There have been published of the Bulletin Vols. L to LIV.; of the- Memoirs, Vols. I. to XXIV., and also Vols. XXVI. to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, XLL, and XLIV. Vols. LV. to LVIII. of the Bulletin, and Vols. XXV., XXX., XXXV., XXXIX., XL., XLIL, XLIIL, XLV. to XLVIII. of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation: — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S. N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission- Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific, in charge of Alexander Agassiz, on the U- S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905, Lieut. Com- mander L. M. Garrett, U. S. N., Commanding. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. Contributions from the Geological Laboratory, Professor R. A. Daly, in charge. These publications are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology^ Cambridge, Mass. Bulletin of the Museum of Ocmpaijfctive Zoology AT HARVARD COLLEGE. Vol. LVIII. No. 5. NOTES ON THE ONTOGENY OF ISOTELUS GIGAS DEKAY By Percy E. Raymond. With Three Plates. CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM. Apuil, 1914, Reports on the Scientific Results of the Expedition to the East- ern Tropical Pacific, in charge of Alexander Agassiz, by the U. S. Fish Commission Steamer "Albatross," from October, 1904, to March, 1905, Lieutenant Commander L. M. Garrett, U. S. Nj Commanding, published or in preparation: — General Report on C. A. AGASSIZ. V.» the Expedition. A. AGASSIZ. I.» Three Letters to Geo, M. Bowers, U. S. Fish Com. A. AGASSIZ and H. L. CLARK. The Echini. H, B. BIGELOW. XVI. i« The Medusae. H. B. BIGELOW. XXIII.s' The Sipho- nophores. H.B. BIGELOW. XXVI.=» TheCteno- phores. R. P. BIGELOW. The Stomatopods. O. CARLGREN. The Actinaria. R. V. CHAMBERLIN. The Annelids. S.F.CLARKE. VIII. « The Hydroids. W. R. COE. The Nemerteans. L. J. COLE. XIX.is The Pycnogonida. W. H. DALL. XIV." The MoUusks. C. R. EASTMAN. VII.' The Sharks' Teeth. 8. GARMAN. XII." The Reptiles. H. J. HANSEN. The Cirripeds. H. J. HANSEN. XXVI1.27 The Schl- zopods. S. HENSHAW. The Insects. W. E. HOYLE. The Cephalopods. W. C. KENDALL and L. RADCLIFFE. XXV. 26 The Fishes. C. A. KOFOID. III.3 IX.» XX.2I) The Protozoa. C. P. R. A. KOFOID and J. R. MICHENER, XXII. 2« The Protozoa. A. KOFOID and E. J. RIGDEN XXIV." The Protozoa. KRUMBACH. The Sagittae. VON LENDENFELD. XXI." The Siliceous Sponges. The Holothuriana. • The Starfishes. The Ophiurans. G. W. MULLER. The Ostracods. JOHN MURRAY and G. V. LEE. XVI 1. 1' The Bottom Specimens. MARY J. RATHBUN. X." The Crus- tacea Decapoda. HARRIET RICHARDSON. II.> The Isopods. W. E. RITTER. IV.« The Tunicates. ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. F. E. SCHULZE. XI.» TheXenophyo- phoras. H. R. SIMROTH. The Pteropods and Heteropods. E. C. STARKS. XIII." Atelaxia. TH. STUDER. The Alcyonaria. JH. THIELE. XV.>5 Bathysciadium. T. W. VAUGHAN. VI.« The Corals. R.WOLTERECK. XVIII. « TheAm- phipods. tBuU. M. C. Z. a Bull. M. C. Z Vol. XLVI., No. 4, April, 1905. 22 pp. VoL XLVI., No. 6, July, 1905, 4 pp.. 1 pi. Vol. XLVI., No. 9, September, 1905, 5 pp.. 1 pi. Vol. XLVI.. No. 13. January, 1906, 22 pp., 3 pis. larv. 1906. 90 do.. 96 dIs. » Bull. M. C. Z , Vol. XLVI., No. 9, September, 1905, 5 pp.. 1 « Bull. M. C. Z., Vol. XLVI., No. 13. January, 1906, 22 pp., 3 6 Mem. M. C. Z., Vol. XXXIII., January, 1906, 90 pp., 96 pis. » Bull. M. C. Z., Vol. L., No. 3, August, 1906, 14 pp., 10 pis. » Bull. M. C. Z.. VoL L., No. 4, November, 1906, 26 pp., 4 pis. »Mem. M. C. Z., Vol. XXXV., No. 1, February, 1907, 20 pp., 15 ' Bull M. C. Z., Vol. L.. No. 6. February, 1907, 48 pp., 18 pis. «> Mem. M. C. Z.. Vol. XXXV, No. 2, August, 1907, 56 pp.. 9 pis. «i Bull. M. C. Z., Vol. LI., No. 6. November, 1907, 22 pp., 1 pi. '2 Bull. M. C. Z., Vol. LIL, No. 1, June, 1908, 14 pp., 1 pL 13 Bull. M. C. Z., Vol. LIL, No. 2, July, 1908, 8 pp., 5 pis. »» Bull. M. C. Z., Vol. XLIIL, No. 6, October, 1908, 285 pp., 22 pis II RiiU AT n 7, Vnl TTT TJn Fi Hrhnhpr 1 QflS 11 nn . 2 nl.s. »* Bull. M. C. Z., vol. JLLiiLi., JNO. 6, uctoDer, it » Bull. M. C. Z., Vol. LIL, No. 5, October, 1908, 11 pp., 2 pis. w Mem. M. C. Z., Vol. XXXVII w^v^r.i^r,. ion » Bull. M. C. Z., Vol. LIL, No. 5, October, 1908, 11 pp., 2 pis. ..,,_. ,, ^ „ ,,.. ,,,.,.,„^^ February, 1909, 243 pp., 48 pis. :i., No. 1, June, 1909, 172 pp., 5 pis., ., Vol. XXXVII., February, 1909, 243 pp.. " Mem. M. C. Z.. Vol. XXXVIIL, No. 1, June, 1909, 172 p » Bull. M. C. Z.. Vol. LIL, No. 9, June, 1909, 26 pp., 8 pis. '» Bull. M. C. Z., Vol. LIL. No. 11, August. 1909, 10 pp.. 3 pis. 3 maps. i» Bull. M. C. Z.. VoL LII M Bull. M. C. Z.. Vol. LIL. No. 13, September, 1909, 48 pp " Mem. M. C. Z., Vol. XLL. August, September, 1910, C. Z.. Vol. LIV.. No. 7. August, 1911, 38 pp. :. O. Z.. Vol. XXXVTTL. No. 2. Dficember. 1' 4 pis. 323 pp., 56 pis. 232 pp., 32 pis. » Bull. M. C. Z.. Vol. LIV.. No. 7. August, 1911, 38 pp. ol. XXXVIII.. No. 2, December. 1911, 232 pp. " Bull. M. C. Z., Vol. LIV., No. 10, February, 1912, 16 pp., 2 pis. V.1 wvTT No. 3, April, 1912. 98 pp. ° -'" *' rsuu. ivi. KJ. Li., vol. JL/iv., INO. lu, rcoruary, lyi:^, id pp., ^ pi 26 Mem. M. C. Z., Vol. XXXV., No. 3, April, 1912. 98 pp., 8 pis, !• Bull. M. C. Z., Vol. LIV., No. 12, April, 1912. 38 pp., 2 pis. " Mem M. C. Z., Vol. XXXV, No. 4, July, 1912. 124 pp., 12 pis, Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE. Vol. LVIII. No. 5. NOTES ON THE ONTOGENY OF ISOTELUS GIGAS DEKAY By Percy E. Raymond With Three Plates CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM. April, 1914. No. 5. — Notes on the ontogeny of Isotelus gigas Dekay. By Percy E. Raymond. Introduction. A partial description of the ontogeny of three of our common asaphids, Isotelus gigas, I. maximus, and Basilicus barrandi, is given on the following pages. The study is based on many hundreds of good specimens, but, as always in palaeontologic work, the material leaves something to be desired, and further specimens will add ma- terially to our knowledge. The important protaspis stages are still missing. The outstanding result of the study is the discovery that an Isotelus gigas, when 3 to 5 mm. long, has almost exactly the same form as an adult specimen of Basilicus barrandi, thus providing an excellent example of recapitulation, for the beginning of the range of Basilicus antedates that of Isotelus. Another interesting fact is that while Isotelus gigas is the most specialized species of the genus, it is one of the first to appear, and apparently one of the first to die out, while the ones which survived to the end of the Ordovician were the more primitive forms, Isotelus maximus and /. iowensis. Two of the three species of the Chazy, Isotelus harrisi and /. platy- marginatus, do not seem to have had any influence on the more persistent and widespread species which belong to the interior faunas. Isotelus harrisi (Ann. Carnegie mus., 1905, 3, p. 343) has a broad flat- tened cranidium, and is not allied to any other species except the Russian /. stacyi, to which Schmidt compared it. Isotelus platy- marginatus (Ann. Carnegie mus., 1910, 7, p. 66) has a very wide de- pressed border on both shields, and is quite unlike any of the later species. The third species, /. arenicola (Ottawa naturalist, 1910, 24, p. 130), is more like I. iowensis or /. gigas, and may have given rise to one or both of those species. While it has the specialized long pygi- dium, the axial lobe is narrow and the genal spines are retained at maturity. 248 bulletin: museum of comparative zoology. IsoTELUS GiGAS Dekay. Plate 1, fig. 1, 2; Plate 2, fig. 2-5; Plate 3, fig. 3. Isotelus gigas Dekay, Ann. Lyceum nat. hist. N. Y., 1824, 1, p. 176, pi. 12, fig. 1. Green, Monthly Amer. journ. geol., 1832, 2, p. 560; Monog. trilobites N. Amer., 1832, p. 68, casts 21, 22. Milne Edwards, Crustac^s, 1840, 3, p. 298. Vanuxem, Rept. Third geol. dist. N. Y., 1842, p. 46, fig. 1. Emmons, Rept. Geol. dist. N. Y., 1842, p. 389, fig. 1. Hall, Pal. N. Y., 1847, 1, p. 231, pis. 60-63. Emmons, Amer. geol., 1855, 1, pt. 2, p. 215, pi. 16, fig. 12. W. B. Rogers, Geol. Penn., 1858, 2, p. 819, fig. 610. Miller, Cincinnati quart, journ. sci., 1874, 1, p. 138. Clarke, Pal. Minn., 1897, 3, pt. 2, p. 701 ipartim) fig. without number. Weller, Pal. N. J., 1902, 3, p. 192, pi. 14, fig. 6, 7 (non 5). Raymond and Narra- way, Ann. Carnegie mus., 1910, 7, p. 53, pi. 15, fig. 1, 2. Raymond, Trans. Roy. soc. Canada, 1912, .ser. 3, 5, sect. 4, pi. 2, figs. 7, 8; pi. 3, fig. 6. Isotelus planus Dekay, Ann. Lyceum nat. hist. N. Y., 1824, 1, p. 178, pi. 13, fig. 7. Green, Monthly Amer. journ. geol., 1832, 2, p. 560; Monog. trilobites N. Amer., 1832, p. 68, cast 23. Isotelus Cyclops Green, Monthly Amer. journ. geol., 1832, 2, p. 560, pi., fig. 7; Monog. trilobites N. Amer., 1832, p. 69, cast. 24. Isotelus megalops Green, Monog. trilobites N. Amer., 1832, p. 70, cast 25. Isotelus stegops Green, Monog. trilobites N. Amer., 1832, p. 71, casts 26, 27. Isotelus jacobus Clarke, Pal. Minn., 1897, 3, pt. 2, p. 706, footnote. ? Asaphus platycephalus Stokes, Trans. Geol. soc. London, 1823, ser. 2, 1, pi. 27, no description. Buckland, Geol. and niin., 1837, 2, p. 73, pi. 45, fig. 12; Ibidem, 1867, pi. 63. Asaphus platycephalus Bronn, Lethaea geogn., 1835, 1, p. 115, pi. 9, fig. 8; Ibidem, 1851-1856, 1, p. 632, pi. 9, fig. 8; pi. 9, fig. 5. Burmeister, Die org. der trilobiten, 1843, p. 127, pi. 2, fig. 12; Ray society edition, 1846, p. 110, pi. 2, fig. 12. Walcott, Bull. M. C. Z., 1881, 8, p. 198, pi. 2, fig. 9. Asaphus gigas Dalman, K. Svensk. vet. akad. Handl., 1826, p. 276; Om Palaea- derna, 1827, p. 91; Ueber die Palaeaden, 1828, p. 70. Asaphus planus Dalman, K. Svensk. vet. akad. Handl., 1826, p. 276; Om Palaeaderna, 1827, p. 91; Ueber die Palaeaden. 1828, p. 70. Asaphus megistos Bilhngs, Geol. Canada, 1863, p. 184, fig. 182. Brongniartia isotelea Eaton, Geol. text book, 1832, p. 33, pi. 2, fig. 12. The following have been erroneously referred to this species: — Isotelus gigas Portlock, Geol. rept. Londonderry, 1843, p. 295, pi. 7, fig. 1; pi. 8, fig. 1. McCoy, Synopsis Silurian foss. Ireland, 1846, p. 53. Brog- ger, Bihang K. Svensk. vet. akad. Handl., 1886, 11, p. 31, pi. 1, fig. 18 (= Isotelus platyrhachis Steinhart, teste Schmidt). Clarke, partim, Pal. Minn. 3, 1897, pt. 2, p. 703, fig. 5 (= Isotelus iowensis). Weller, Pal. RAYMOND: ISOTELUS GIGAS DEKAY. 249 N. J., 1902, 3, pi. 14, fig. 5, cop}' of preceding (= Isolclus iowensis). Grabau and Shimer, N. A. index fossils, 1910, 2, p. 293, fig. 1()02 (= Iso- ielus latus Raymond). Asaphus {Isotelus) gigas Salter, Mem. Geol. surv.. Unit. Kingdom, IStil, dec. 11, pi. 3; Mon. Brit. Silur. tril., 18G5, p. 161, pi. 24, fig. i-5; ? pi. 25, fig. 1. Reed, Lower Silur. tril. Girvan dist., 1904, p. 45, pi. 7, fig. 1. Asaphus gigas ? Nicholson and Etheridg(\ Mon. Silur. foss. Girvan di.st., 1879, fa.sc. 2, p. 153, pi. 10, fig. 18, 19; Mem. Geol. surv. Silur. rocks Brit., 1899, 1, Scotland, p. 509, 513 514 (= Isotelus instabilis Reed). Asaphus platycephalus Nieszkowski, Arch, naturk. Liv.- Est.- und Kurl., 1857, ser. 1, p. 551, pi. 1 ( = Isotelus remigerus Eichvvald and /. 7-obustus Roemer, teste Schmidt). Billings, Geol. Canada, 1863, p. 184, fig. 183 (= Iso- telus latus Raymond) ; Cat. Silur. foss. Anticosti, 1866, p. 24, fig. 7 ( = Iso- telus latus Raymond) ; p. 26, fig. 8b (= Brachyaspis altilis Raymond); Quart, journ. Geol. soc. London, 1870, 26, p. 486, pis. 31, 32 (= Isotelus latus Raymond). From the above synonymy, text-book and catalogue references have for the most part been omitted, as have also references where there is neither description nor original figure. So far as can be judged from the pubUshed figures and descriptions, none of the foreign specimens referred to this species really belong to it, and most of those which have at one time or another been so referred are now known by other names. Ontogeny, In a recent paper (Ann. Carnegie mus., 1910, 7, p. 53), Mr. Narra- way and the writer summarized the ontogeny of Isotelus gigas. The Walcott collection in the M, C. Z. contains more complete material than we then had, and permits a study of the species from the stage in which it had a length of only 3 mm. The smallest specimen (No. 36) in the collection is 3 mm. long, and has the same width at the genal angles. It is exposed from the lower side, and retains the hypostoma in position. The details of the thorax and pygidiura are not well shown, but the cephalon occupies at least one half the length, and the pygidium is considerably smaller. The genal spines are long, extending back nearly to the posterior end of the pygidium (Fig. 2). Another specimen, 5 mm. long, is also exposed from the lower side, and shows the hypostoma. In both, the hypostomas are deeply cleft behind, and rather fiat, without the convex body of the hypostoma of either Isoteloides or Asaphus. The smallest specimen (No. 45) showing the dorsal aspect of the 250 bulletin: museum of comparative zoology. shell is 3.5 mm. long as it lies in the matrix, but as the pygidium is somewhat bent under the body, the actual length is probably .5 mm. greater. The cephalon is 2 mm. long and 4 mm. wide (semicircular) and is bordered by a wdde (.5 mm.) concave brim. The glabella is convex, abruptly elevated above the brim at the front, bounded at the sides by deep dorsal furrows which converge backward. Between the eyes the glabella is marked by a pair of deep glabellar furrows. Be- hind them, on the narrowest part of the glabella, is a median tubercle, and on each side at this point is an isolated basal lobe. In short, the glabella is like that of an adult specimen of Basilicus (Plate 1, fig. 1). The facial suture can not be made out with absolute certainty on these smallest specimens. On a specimen with the cephalon 1.75 mm. long it seems to be marginal in front, while on a finely preserved cephalon 4 mm. long (No. 37), it is certainly intramarginal (Fig. 1). The eyes in the smallest specimens are very large, and only about one half their own length from the posterior margin. On a cephalon 2.25 mm. long, the eyes are 1 mm. long and each forms a semicircle, the palpebral lobes almost touching the glabella in front. The posterior corner of the eye is only .3 mm. from the posterior mar- gin of the cephalon. The eye is there- fore relatively much larger than in the adult, but no further forward. The growth after this stage is more rapid in front of the eyes than behind them, and a cephalon 3 mm. long has the eyes still only 1 mm. long. The thorax of the smallest speci- mens shows feW' peculiarities other than the narrowness of the axial lobe. The specimen which is 3 mm. long has the axial lobe .5 mm. wide, or .2 the total width of the thorax. The smallest specimen with a well-preserved pygidium is 3.5 mm. long, and the pygidium is 1.25 mm. long and 2.5 mm. wide. The axial lobe is very con\'ex, narrow, and is prominent all the way to the end. There is a narrow concave border, and the axial lobe reaches to and overhangs this border. The pleural lobes are convex, somewhat triangular, and crossed by four pairs of furrowed ribs. The axial lobe has one strong, and two weak rings at the anterior end. Another Fig. 1. Isotelvs gigas Dekay. A small specimen in the Ogygites stage, the facial sutures being intramarginal. The pygidium is incurved so that the posterior part is not well shown. X 6. M. C. Z. No. 37. RAYMOND: ISOTELUS GIGAS DEKAY. 251 pygidium, 2 mm. long, shows the furrowed ribs, but less strongly, and pygidia 3 mm. long are quite smooth, except for the rather promi- nent axial lobe. The pygidia of young specimens are much smaller than the cephala, being both shorter and narrower. Summary of characters of specimens less than 5 mm. long. The cephalon makes up one half the entire length, and has a greater area than all the remainder of the shell ; it is semicircular in outline, with a wide concave brim, and long genal spines. The eyes are very large, close to the glabella and to the posterior border. The glabella is convex, prominent at the front, narrow behind, and expanding for- ward, with a median tubercle between the isolated lobes at the posterior end. The free cheeks apparently do not meet in front. The thorax consists of two or three segments; the axial lobe is less than one fourth the total width. The pleura are flat to the fulcrum, which is far out, and are there turned down abruptly. Each pleuron bears a straight furrow, which extends beyond the fulcrum. The pygidium is small, semicircular, with narrow, prominent axial lobe, and narrow, concave border. The pleural lobes are crossed by distinct, furrowed ribs, and there are rings on the anterior end of the axial lobe. Changes duriyig development. — It will be shown in the sequel that some of these youthful characters are lost much more quickly than others. The first to be changed is the marginal position of the facial suture, and the most persistent is the genal spine. The ribs on the pygidium disappear very early, while new segments are still being added to the thorax. Then the glabella becomes flattened, the axial lobe becomes wider, the eyes relatively smaller, the genal spines shorter, and the cephalon more triangular and smoother. Outline of cephalon and pygidium. — The outlines of the shields, and their changes, were quite fully discussed by Raymond and Xarraway, but the present more complete material has furnished some additional facts. In the smallest specimens so far seen, both the cephalon and pygidium are one half as long as wide, and regularly semicircular. In the cephalon, this condition persists until this shield reaches a length of about 4 mm., when the ratio of length to breadth begins to rise, and at the same time the sides of the cephalon become straighter, so that the general form becomes subtriangular, instead of semicircu- lar. This change takes place rapidly, so that a cephalon 12 mm. long is three fourths as long as wide, and few specimens of any size have a higher ratio than this. Most cephala whose length is between 12 mm. and 30 mm. have this ratio (length divided by width), between .72 252 bulletin: museum of comparative zoology. and .80. In specimens with cephala over 30 mm. long, the growth seems to be more rapid in the transverse direction, and the ratio drops again. It is only .62 in the two largest specimens in the collection, the largest cephalon being 60 mm. long. The form in these large specimens remains subtriangular, and does not return to the semi- circular outline of the young. Pygidia show a similar series of changes, but the index rises higher, and does not show so great a falling off in large specimens. In the adults, the pygidia are usually of about the same length, or a little longer, than the cephala, but are always narrower, hence the higher index. In specimens where the length of the cephalon is .75 the width, the length of the pygidium is generally .80 the width. In the larger specimens, where the indices of both shields have begun to drop, the difference is more noticeable, the index of the cephalon being .62 while the index of the pygidium is .77 in one well-preserved specimen. Raymond and Narraway found occasional specimens in which the length equaled the width, but in the present collection the highest index noted was .85. In small specimens, the pygidium is about one half as long as the cephalon, and the two shields do not reach quite equal length until the whole animal has attained a length of about 50 mm. In specimens more than 100 mm. long there is a tendency for the pygidium to exceed the cephalon in length, the difference being as much as 10 mm. in some large specimens. That the pygidium does, in the adult, finally reach the same length as the cephalon, seems to be due, not to an acceleration in growth at any particular time, but rather to a retarda- tion of the longitudinal growth of the cephalon, during the process of widening. The pygidium never becomes quite as wide as the cepha- lon, so that more energy can be put into longitudinal growth. Glabella and axial lobe of the pygidium. — The glabella soon loses its convexity and prominence, as well as its furrows, basal lobes, and median tubercle. These features are still visible on a cephalon 4.5 mm. long, but are faint, and the front of the glabella no longer rises abruptly from the brim. The outline of the glabella is still distinguishable on a cephalon 7.5 mm. long, but it is there hardly more convex than in the adult. Something of the form of the glabella can be made out on most specimens, whatever the size, but it is rather smoothly merged into the general surface of the cephalon in all specimens with a cepha- lon more than 8 mm. long. The axial lobe of the pygidium remains convex and conspicuous somewhat longer than the glabella does. On a specimen 9.5 mm. RAYMOND: ISOTELUS GIC, AS DKKAY. 253 long, where the glabella is quite flat, the axial lobe of the pygidium is prominent, even at the posterior end, which rises abruptly from the flattened boi-der. It would be expected that the pygidium would retain its youthful characters longer than the cephalon, for it is really younger, and has had less time to change. On a specimen 18 mm. long, (pygidium 5.5 mm. long), the axial lobe is fairly prominent, but does not extend quite to the border, and is less convex throughout its length than in the smaller specimens. It is of course, more or less distinguishable on specimens of all sizes, but is very faint on pygidia more than 25 mm. long, except under certain conditions of preserva- tion. The axial lobe also grows shorter as the shell increases in length. In young specimens it reaches and overhangs the flattened border while in large individuals the posterior end is at a distance in front of the border equal to the width of the border itself. Border. — The border on the cephalon and pygidium, which is very wide in the young specimens, becomes much na,rrower with further growth. In very young specimens its width is equal to one fourth the length of the cephalon, but in the adult it is only about one sixth the length. In the young the plane of the brim is horizontal, but in older specimens it becomes gradually more inclined and less concave. This fact has an important bearing upon the relative primitiveness of the asaphids with a border and those without. In a previous paper (Trans. Roy. soc. Canada, 1912, ser. 3, 5, sect. 4, p. HI), I have con- sidered the absence of a depressed border to be primitive in this family, but evidence seems to be accumulating that the contrary is the case. Beside the al)ove, one may cite the condition seen in certain species of Onchometopus, especially 0. simplex, and an undescribed form from the Eden at Cincinnati. In these species, the pygidia, which, as has just been shown, have a tendency to lag behind the cephala in develop- ment, often have a very decided trace of a depressed border, as though they may have been developed from an Isotelus-like form. Further, all the strongly segmented asaphids, like Basilicus, Ogygopsis, Ogygio- caris, and Ogygites, have more or less of a border, and it is only the rather smooth forms, like Nileus, Asaphus, Onchometopus, and the like, which lack it. Genal spines. — The reduction of the genal spines during life was discussed at length by Clarke (Pal. Minn., 1897, 3, pt. 2, p. 704), and the present collection confirms the previous statements. The smaller specimens, however, show that the spines of the young were even longer than was supposed, for specimens 3.5 to 5 mm. long bear spines which extend back as far as the posterior end of the p\gidium. The 254 bulletin: museum of comparative zoology. smallest specimens do not seem to have the spines quite so long, but this may be an accident of preservation. A specimen 8 mm. long has spines extending to the front of the pygidium, and one 18 mm. long has them reaching the sixth segment. Specimens about 30 mm. long show spines extending to the fourth segment, and those 35 to 50 mm. long show small, narrow spines reaching the third segment. Speci- mens above 55 mm. long do not show any spines at the angles, except in the case of a few specimens, which retain them as youthful charac- ters are sometimes retained by adults of any species. Thorax. — The full number of segments is attained when the animal is about 8 or 9 mm. long, and the following are measurements of speci- mens with less than 8 segments : — 2 segments 3 segments 4 segments 5 segments 6 segments 7 segments 8 segments specimen, 3 mm. long specnnen, 6 mm. long: 2 specimens, 3.5 and 5 mm. long, respectively: 1 specimen, 5.5 mm. long: 1 specimen, 7 mm. long: 1 specimen, 8 mm. long: 1 specimen, 6.5 mm. long: smallest specimen, 9.5 mm. long. In the smallest specimens the axial lobe of the thorax is very narrow, being only one fifth the total width in the specimen which is 3 mm. long. A specimen 5 mm. long has a thorax of three segments, and the axial lobe is one fourth the width of the thorax. In a specimen 8 mm. long this ratio has risen to .30 and one 10 mm. long has the axial lobe .40 of the width of the thorax, while a specimen 18 mm. long has the same index. Specimens 29 and 32 mm. long have the index .46, and one 38 mm. long has the axial lobe one half the total width of the thorax, which is the normal ratio for adults of this species. The exact ratio .50 was found in eleven specimens varying in length from 38 to 130 mm. and the greatest departure noticed in all the specimens measured was .05, the ratio varying from .45 to .53, the latter index being found in a specimen 181 mm. long. Summary of the mitogcny. — Basilicus stage. The glabella is convex, narrowed behind, with a median tubercle and basal lobes. The brim is wide and horizontal, the cephalon semicircular in outline, the genal angles produced into long spines. The axial lobe of the thorax is narrow, the pygidium has a prominent axial lobe, and the pleural lobes of the pygidium are ribbed. Ogygites stage. The glabella soon becomes flattened, and the facial sutures intramarginal, as in the adult of Ogygites. RAYMOND: ISOTELUS GIGAS DEKAY. 255 IsoTELrs stage. The cephalon and pygidium both become smooth, and the axial lobe widens, as in the adult of Isotelus. IsoTELUs GIGAS Stage. The cephalon and pygidium become tri- angular, and the spines are lost from the genal angles, a combination of characters distinguishing this species. ^3 Fig. 2. — A very small specimen of Isotelus gigas, exposed from the imder side. This specimen shows the rounded cephalon and pygidium and divided hj-postoma, though it is too poorly preserved to show how many thoracic segments are present. X 10. ' M. C. Z. No. 36. Fig. 3-7. — The same species. A series of specimens showing the changes from rounded shields, narrow axial lobe, and long genal spines of the young to the pointed shields, wide axial lobe, and spineless cheeks of the adult. Fig. 3, 4 represent an Isotelus maximus stage, 5 and 6 an I. iowensis stage and 7 is as small a specimen as is often found showing all the characteristics of the adult of /. gigas. Natural size. M. C. Z. Nos. 48, 38-41. Isotelus iowensis Owen. Plate 2, fig. 6; Plate 3, fig. 1, 2. Isotelus iowensis, Owen, Rept. Geol. sur. Wise, Iowa, Minn., 1852, p. 577, pi. 2a, figs. 1-7. Clarke, 1897, Pal. Minn., 1897, 3, pt. 2, p. 704. Although mentioned occasionally in faunal lists, this species seems to have been pretty generally neglected. It is so closely allied to Isotelus gigas, that, where the two species occur together, as at Trenton Falls, it seems almost like hair-splitting to recognize two species. Essentially, it is an Isotelus gigas which retains at maturity certain 256 bulletin: museum of comparative zoology. youthful characters usually lost by that species. Thus, the adult Isotelus iowensis has long genal spines, extending to the fifth, sixth, or seventh segment, an axial lobe a little less than half the width of the thorax (.41 to .46), and pygidia show a fairly convex axial lobe, and traces of ribs on the pleural lobes. The pygidia are also somewhat more rounded than those of Isotelus gigas, although fully as long. As minor features of the species, it may be noted that the anterior portion of the glabella is quite well defined, and the geniculation of the pleural lobes of the pygidium is further from the dorsal furrows than in Isotelus gigas. The collection contains a number of cranidia and pygidia, and two complete specimens from the Maquoketa at Elgin, Iowa, and a single complete specimen which is presumably from Iowa but not labeled as to locality, so that direct comparisons can be made with the specimens from Trenton Falls, and the specimens from the eastern and western localities are found to agree in all particulars. The specimens from Trenton Falls have usually been identified as Isotelus maximum. Two specimens from the Low\'ille and Black River at Ottawa, Ontario, described by Raymond and Narraway (Ann. Carnegie mus., 1910, 7, p. 56, pi. 15, fig. 3) as Isotelus sp. ind. probably belong to this species. They were separated from Isotelus gigas because they had more rounded pygidia with a rather prominent axial lobe, and they have a rather narrow axial lobe in the thorax. In 1910 there was no opportunity of comparing the specimens from Ontario with Isotelus iowensis, but I now find that the^^ agree very closely with the speci- mens from Iowa. The specimen from the "Hudson River" at Granger, Minnesota, figured by Dr. Clarke ^ as a specimen of Isotelus maximus, is, as figured, an excellent example of /. iowensis, the shields being too long and narrow for a typical specimen of /. 7uaximus. The t^pe of this species, which is now in the collection at the Walker Museum of the University of Chicago, was obtained from the IMaquo- keta shale, which is of Upper Ordovician (Richmond) age, according to the most recent correlations. The specimens from the Trenton, though sometimes larger, are too much like the Iowa specimens to be distinguished as a distinct species at the present time. Additional note: — Since this paper was written Mr. A. W. Slocum's excellent account (Field mus. nat. hist., publ. 171, Geol. ser. 4, no. 3, p. 48, pi. 13, fig. 1, 2) of the trilobites of the Maquoketa beds of Fayette > PaL Minn., 1897, 3, pt. 2, p. 703, fig. 5; also reproduced by Weller as Isotelus gigas, Pal. N. J., 1902, 3, pi. 14, fig. 5. RAYMOND: ISOTELUS GIGAS DEKAY. 257 Co., Iowa, has appeared. He describes and figures Isotelus iowensis from specimen obtained at the type locality. Mr. Slocum figures (PI. 13, fig. 2) one remarkable specimen which has the appearance of possessing an epistome, and he has described the sutures as indi- cating the presence of this plate. Mr. Slocum was kind enough to show me this specimen, and while there are certain lines which have decidedly the appearance of epistomal sutures, I am unable to believe that this species has an epistome, that plate being entirely unknown among the Asaphidae. The facial suture is described by Slocum as marginal, but although it is nearly so, many specimens show clearly that it is intramarginal or "Isoteliform." The M. C. Z. contains two specimens from the Moquoketa of Iowa which show the doublure of the cephalon. Both show a median verti- cal suture such as is seen in all other Asaphidae and one of them (M. C. Z. No. 435) shows at the left a suture such as that on Mr. Slocum's specimen. There is no corresponding one on the right side, however, and the other specimen (M. C. Z. No. 442) shows no trace of such a suture on either side. Just what these lines indicate is not at present evident. Isotelus maximus Locke. Isotelus rnaximns Locke, Second aiin. rept. Geol. surv. Ohio, 1838, p. 246, fig. 8, 9. Clarke, Pal. Minn., 1897, 3, pt. 2, p. 701 (not fig. 5-7). Raymond and Narraway, Ann. Carnegie mus., 1910 7, p. 55, fig. 3. Isotelus megisios Locke, Trans. Amer. assoc. geol. and nat., 1841-, p. 221, pi. 6; Amer. journ. sci., 1842, ser. 1 42, p. 366, pi. 3. Meek, Pal. Ohio, 1873, 1, p. 157, pi. 14, fig. 13. Miller, Cmcinnati quart, journ. sci., 1874, p. 137. Asaphus megistos Walcott, Science, 1884, 3, p. 200, fig. 1. Typical and well-preserfed specimens of this species from about Cincinnati are very easily distinguished from specimens of Isotelus gigas from Trenton Falls, or specimens of Isotelus iowensis from Iowa, but they are not always so easily separated from the other two species of Isotelus found with them at Cincinnati. The chief reason for the confusion which has arisen as to the characteristics of the two species, Isotelus gigas and /. maximus, is that five species, belonging to two genera, have been identified under these two names. At Cincinnati, the common asaphids are Isotelus gigas, I. maximus, I. latus, and un- described species of Isotelus and of Onchometopus. At Trentoii Falls the only species are Isotelus gigas and /. iowensis, but the /. iowensis 258 bulletin: museum of comparative zoology. there has been called /. maximus. As /. iowensis has certain char- acters half-way between /. gigas and I. maximus, this misidentification has served to throw the Cincinnati specimens into hopeless confusion. The following key will show the more important characters which separate the genera of asaphids occurring in the formations from the Chazy to the Richmond. A. Depressed or concave border on both shields. a. Surface of shields ribbed. 1. Free cheeks meeting in front Ogygites. 2. Free cheeks separated in front Basilicus. b. Surface of shields not ribbed. 1. Axial lobe narrow, prominent Isoteloides. 2. Axial lobe wide, depressed Isotelus. B. Concave border on pygidium only. a. Eyes very large Nileus. b. Eyes small but very high Vogdesia. C. No concave border on either shield. a. Free cheeks meeting in front Onchometopus. b. Free cheeks separated in front Brachyaspis. The four more common species of Isotelus in the Middle and Upper Ordo vician can be separated as follows : — A. Shields about three fourths as long as wide. a. Adult without genal spines. Shields subtriangular . ./. gigas. b. Adult with genal spines. Shields rounded /. iowensis. B. Shields less than three fourths as long as wide, regularly rounded. a. Adult without genal spines /. latus. b. Adult with genal spines I. maximus. This may at first sight seem to be an arbitrary arrangement of the species, but these are not the only characteristics in which the above species differ, the other points being given in the general discussion of each species. The geographical distribution is also in agreement with the above separation. For instance, the specimens which I have described as Isotelus latus were considered by Billings the female of /. gigas, but if this is true, why are all the hundreds of specimens found at Trenton Falls, the narrow form? I have measured an extensive series of each species, and find the ratio of length to the width of the shields to be an important clue to their identification. From the above key, it might seem that it would RAYMOND: ISOTELUS GIGAS DEKAY. 259 be difficult to distinguish pygidia of Isotelus giga-f(, I. maximus, and /. iowensis, but such is not the case. If the pygidium has straight sides, it can quickly be placed as I. gigas. If the posterior end is rounded, it might be taken for either /. iowensi'} or /. maxitnus, but if the ratio of length to breadth is above .65 it is probably /. iowen- sis, and if below, /. maximum. In all cases, however, one must be dealing with adult specimens, and must take all characteristics into account. With isolated yoimg specimens, it is not always possible to state to what species they belong, as the specific characters are not assumed until late in life. A case in point is the young specimen described by Meek as Pro'etus spurlocki (Pal. Ohio, 1873, 1, p. 161, pi. 14, fig. 12). Proetus spurlocki Meek. The type of this species (Plate 1, fig. 3) which is in the Dyer collec- tion in the M. C. Z., is clearly a young Isotelus, but whether it is the young of Isotelus gigas or of /. maximus, the writer is not able to decide. The specimen is 8.5 mm. long, and when compared with a specimen of the same size from Trenton Falls, the only apparent difference is that the specimen from Cincinnati has only seven thoracic segments, while the Isotelus gigo^ has eight. Both have long cheek spines, small pygidia, rather long cephalons, and narrow axial lobe. The ratio of length to ^N-idth in the cephalon of the "Proetus" is rather high, (.69), for a specimen of Isotelus gigas of this size, but it is also much higher than one would expect in /. maximus. The smallest specimen in the collection which is surely identifiable as Isotelus maximus is consider- ably larger than the type of P. spurlocki, being 16 mm. long. In this specimen, the length of the cephalon is .62 of the width. This is slightly above the average for adult specimens, where this ratio ranges from .57 to .60. The specimens of Proetus spurlocki which have been found at Cincinnati seem to have all come from the Eden shale, where Isotelus' maximus seems to be more common than /. gigas, and on that ground the presumption would be that this specimen belongs to the former species. It seems very probable, in any case, that the young of the two species would be alike at this stage of development. Changes during the life history of Isotelus maximus. — Isotelus maxi- mus seems to be a much less variable form than /. gigas. Being, for the genus, a relatively primitive form, it reaches its specific habit quite early, and the principal variations among the specimens more 260 bulletin: museum of comparative zooLOoy. than 20 mm. long are in the matter of size and the shortening and thickening of the genal spines. Nearly all the specimens in the Dyer collection are either incomplete or enrolled, so that the specimens have to be considered, not in terms of their total length, but in accordance with the length of one of their shields. The smallest specimen in the collection has a pygidium 6 mm. long, and the largest pygidium is 128 mm. long. This species differs from I. gigas in that the two shields become equal at an early age, and stay equal throughout the greater part of the adult stage. One adult specimen, (cephalon 46 mm. long), has the pygidium shorter and more nearly semicircular than the cephalon, which is the reverse of what is found in /. gigas. There are not enough good specimens to show whether this is the general rule in this species. According to the measurements of thirteen specimens from Cincinnati, the length of the cephalon averages a trifle less than .6 the width, and the pygidium is usually a little longer, the average being .64. The axial lobe of the thorax averages about .42 of the total width, and there is surprisingly little variation, the extremes being .40 and .44. The axial lobe is then generally a little narrower than in /. gigas. Isoiclus maximus from Toronto, Ontario.^ Thvough the courtesy of Prof. W. A. Parks, I have been able to study a series of very fine speci- mens in the collection at the University of Toronto. The specimens were all from the Lorraine in the vicinity of Toronto, and were very well preserved, though generally a little flattened. So far as the writer could determine from the large collection at the University, and from a short experience in the field, Isotchis maximtis is the only asaphid present in the Lorraine at Toronto. The specimens are quite large, ranging from 70 to 285 mm. in length, and they are extended, not enrolled as is generally the case with the specimens from Cincin- nati. They show remarkably short, wide cephalons, the average ratio of length to width of 11 cephalons, 9 of which were over 50 mm. long, being .46, and the range of variation, .43 to .51. The pygidia are in all cases longer than the cephalons, and the ratio for fifteen specimens averages .59. The axial lobe of the thorax also averages a little narrower than in specimens from Cincinnati, the average of fifteen specimens being .374, and the limits of variation .34 to .40. The largest specimen which was well enough preserved to yield accurate measurements was 262 mm. long. The cephalon made up .30 of the length, the thorax .33, and the pygidium .37. The greatest width was .69 of the length. The Dyer collection contains a poorly preserved specimen from INIorrow, Ohio, seven of whose thoracic RAYMOND: ISOTELUS GIGAS DEKAY. 261 segments together measure 112 mm. in length. If this specimen had the same proportions as the one from Toronto, it would, when com- plete, have been 384 mm. long, and 260 mm. wide at the genal angles, thus giving a very large surface area. Basilicus barr.\ndi (Hall). Plate 1, fig. 4, 5; Plate 2, fig. 1, 7. Asaphus barrandi Hall, Foster and Whitney Rept. Lake Huperior land dist., 1851, pt. 2, p. 210, pi. 27, fig. 1, a-d; pi. 28. Geol. Wise, 1862, 1, p. 41, fig. 4. Asaphus romingeri Walcott, 28th ann. rept. N. Y. state mus., 1879, p. 78. Asaphus wisconsensis Walcott, Ibid., p. 79. Plychopyge ulrichi Clarke, Pal. Minn., 1897, 3, pt. 2, p. 709, figs. 12, 13. Basilicus romingeri Raymond and Narraway, Ann. Carnegie mus., 1910, 7, p. 49, pi. 15, fig. 9, 10; pi. 16, fig. 1^. When Raymond and Narraway wrote, they did not have access to the Foster and Whitney report, or they would probably have adopted Hall's name for this rather common Black River Basilicus. In view of the remarkable resemblance of the young of Isotelus gigas to the adult of this Basilicus, it deserves to be more adequately figured than it has been hitherto. The species seems to be of wide geographic and narrow vertical range, and should be better known than it is. Hall's types of Asaphus barrandi consisted of an entire specimen with the cephalon mutilated and showing the hypostoma in place, two imperfect pygidia, a free cheek, and a good pygidium. These speci- mens are preserved in the American museum of natural history, where I have had the opportunity to study them, through the courtesy of Dr. E. O. Hovey, Curator of Geology. Four of the specimens, (Hall, Loc. cit., pi. 27, fig. lb, c, d, and pi. 28), including the entire specimen, are from Platteville, Wisconsin, and one, (pi. 27, fig. la), is from St. Joseph Island, Ontario. Platteville may, then, be con- sidered the type locality for the species. Asaphus romingeri and .4. wisconsensis were described by Walcott without illustration, but are represented by a considerable number of fragmentary specimens in the Walcott collection of the M. C. Z. Two imperfect cranidia from the Black River at Russia, Herkimer county, N. Y., have attached to them original labels indicating that they are the types of the two species. These labels have, however, 262 bulletin: museum of comparative zoology. been interchanged, for the specimen having the wide border in front is labeled A. rovvingcri, while the description states that A. iviscon- sen^is has the wide border. With the exception of this single char- acter, the width of the flat border in front of the glabella, the two types are in perfect agreement. Raymond and Narraway, without seeing the types, hazarded the opinion that Walcott's two names represented only one species, and cited the fact that in their collections, the border was wider in young specimens than in older ones. Jt is true that the type of AsapJms ronmigcri is larger than that of A. wisconscnsls, but the other specimens in the collection fail to bear out our suggestion. Beside the type, there is only one other specimen which will answer the definition of Asaphus romingeri, and that is somewhat smaller than the type of A. unsconsensi^, while there are several specimens of A. wisconscnsis which are larger than the type of A. romingeri. This difference in the width of the border certainly does exist among adult specimens, and while it does not seem to be of specific value, it is a character which may later prove of use. Clarke's types of Ptychopyge ulrichi were two pygidia from Cannon's Falls, Minnesota, and they agree with pygidia from Platteville, Wis- consin, and Newport, New York, except that a complete pygidium of the same size as Clarke's large fragment shows a much shorter form than is indicated by his outline restoration. The collection of the M. C. Z. contains, beside the specimens in the Walcott collection, which are all from Russia (Newport), New York, a number of cranidia and pygidia, and one nearly complete specimen, from Platteville, Wisconsin. The specimens in the Walcott collection were presumably all used in describing Asaphus romingeri and A. wis- consensis, although there are pygidia present, a fact not mentioned in the original descriptions. The nearly entire specimen is flattened, and is somewhat imperfect in several particulars, but shows fairly accurately the general propor- tions of the animal. The cephalon is large, occupying .43 of the length of the animal, while the thorax occupies .27, and the pygidium .30. The axial lobe is narrow, .32 of the total width, and the cephalon and pygidium are both nearly semicircular, the length of the cephalon being .55 of the width, and the length of the pygidium .53 of the width. These general proportions accord very well with a specimen of Isotelus gigas 3.5 mm. long, except that the thorax being complete in the Basilicus, it makes the whole animal longer. Separated cranidia from St. Joseph Island, Lake Huron, (in collec- tion of the Geological survey of Canada), from Ottawa, Ontario, RAYMOND: ISOTELUS GIGAS DEKAY. 263 (in collection of J. E. Narraway, Esq.), from Faribault, Minnesota, (in collection of the Carnegie museum), from Platte ville, Wisconsin, and from Newport, New York, have been compared, and measurements taken to determine the ratio of the length of the glabella to the total length of the cephalon. It was found that the glabella was relatively shorter on young specimens than on older ones, although there was not much difference. A cranidium 7 mm. long has the glabella .78 of the total length, while a specimen 62 mm. long has the glabella .88 of the length. The width of the brim is therefore relatively only about half as great in the old specimen as in the young one. On the adult specimen of A. wi^conscnsw in which the brim is widest, it occupies .19 of the length, and on the specimen on which it is narrowest it occupies only .12, the average being .15. On the t}T)e of A. romingeri it occu- pies onl}^ .07 of the length, and on the only other specimen of this sort, it occupies .11. With the exception of these two specimens, cranidia from all the localities conform closely to the general average. The agreement in proportions and ribbing among the pygidia from the localities just mentioned is remarkable. The proportion of length to width is quite constant in specimens above 20 mm. long, and is usually about .60. The ribbing on the larger specimens is only a little less strong than on the small ones, and all have the same number of ribs and rings. The largest pygidium in the collection is from Platteville, and is 64 mm. long and about 105 mm. wide. Katmond. — Isotelus gigas Dekay. EXPLANATION OF PLATES. PLATE 1. Fig. 1. — Isotelus gigas DeKay. An immature specimen with (apparently) three segments in the thorax. From the Trenton at Trenton Falls, N.Y. X 15. No.45, M.C. Z., WalcottcoU. Fig. 2. — The same species. A somewhat larger but still immature individual showing the short rounded pygidium of the young. Thorax partially buried in the matrix. From the Trenton at Trenton Falls, N. Y. X 6. No. 48, M. C. Z., Walcott coll. Fig. 3. — Isotelus maximus ? Locke. The type of ProUus spurlocki Meek. From the Eden at Cincinnati, Ohio. X 6. No. 43, M. C Z., Dyer coll. Fig. 4. — Basilicus barrandi (Hall). A specimen with a narrow brim, similar to the type of Asaphus romingeri Walcott. From the Leray- Black River at Newport, N. Y. X |. No. 46, M. C. Z. Walcott coll. Fig. 5. — The same species. A nearly complete but flattened specimen some- what restored about the eyes and at the posterior end of the pygi- dium. From the Black River at the quarry on Limestone Creek, near Platteville, Wise, the type locality of the species, xf - No. 34, M. C. Z., Whitney coll. Bull. Mus. Comp. Zool. Isoteliis Plate i / HELIOTYPE CO., BOSTON Raymond. — Isotelus gigas Dekay PLATE 2. Fig 1. — Basilicus harmndi (Hall). The imperfect cranidium which was the type of Asaphus wisconsensis Walcott. From the Leray-Black River at Newport, N. Y. X |. No. 47, M. C. Z., Walcott coll. Fig. 2-5. — Isotelus gigas Dekay. Young specimens in various stages of growth, illustrating the change from the rounded to the pointed pygidium and the shortening of the genal spines. From the Trenton at Trenton Falls, N. Y. X |. No. 48, 38, 39, 40, M. C. Z., Walcott coll. Fig. 6. — Isotelus iowensis Owen. An imperfect specimen from an unknown locality in Iowa. To be compared with a young Isotelus gigas from Trenton Falls (fig. 5) and with a young /. iowensis from the same locality (Plate 3, fig. 2). X |. No. 420, M. C. Z., ? Sir Charles Lyell coll. Fig. 7. — Basilicus harrandi (Hall). The specimen which was the type of Asaphus romingeri Walcott. From the Leray-Black River at Newport, N. Y. X f . No. 35, M. C Z., Walcott coll. Bull. Mus. CoMip. Z()(")l. Isolt-llis Pl;i!f 2 HELIOTYPE CO., BOSTON Raymond. — Isotelus gigas Dekay. PLATE 3. Fig. 1. — Isotelus iowensis Owen. A large specimen from the Trenton at Trenton Falls, N. Y. 7 mm. longer than natural size. No. 422, M. C. Z., Walcott coll. Fig. 2. — The same species. A small specimen from the same locality as the last, for comparison with fig. 3, and with fig. 6, Plate 2. No. 421, M. C. Z., Walcott coll. Fig. 3. — Isotelus gigas Dekay. A small specimen showing all the character- istics of the adult. For comparison with fig. 2. From the Tren- ton at Trenton Falls, N. Y. No. 41, M. C. Z., Walcott coll. Bull. Mus. Comp. Zool. Isotelus Plate 3 HELIOTYPE CO., BOSTON The following Publications of the Museum of Comparative Zoology are in preparation: — LOUIS CABOT. Immature State of the Odonata. Part IV. E. L. MARK. Studies on Lepidosteus, contiuued. " On Arachnactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II.. with 14 Plates. H. L. CLARK. The "Albatross" Hawaiian Echini. Reports on the Results of Dredging Operations in 1S77, 1878, 1879. and 1880. in charge of Alexander Aqassiz, by the U. S. Coast Survey Steamer " Dlalce," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. The Crustacea of the "Blake." A. E. VERRILL. The Alcyonaria of the "Blake." ■Reports on the Results of the E.\p'edltion of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tanner, U. S. N., Cominanciirig, in charge of Alexander Agassiz, as follows: — K. BRANDT. The Sagittae. The Thalassicolae. ■O <:ARLGREN. The Actinarians. B. V. CHAMBERLIN. The Annelids. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes of Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. " The Schizopods. HAROLD HEATH. Solenogaster. W. A. HERDMAN. Tlic Ascidians. S. J. HICKSON. The Antipatliids. E. L. MARK. Brancliiocerianthus. JOHN MURRAY. The Bottom Specl- mens. P. SCHIEMENZ. Tiie Pteropods and Heteropods. THEO. STUDER. The Alcyonarians. The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. R. H. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser. U. S. N., Com- manding, as follows: — MARY J. RATHBUN. The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes. T. W. VAUGHAN. Tlie Corals, Recent and Fossil. V. CHAMBERLIN. . The AnneUds. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. •S. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. The Ophiurans. •G. W. MULLER. The Ostracods. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. There have been pubHshed of the Bulletin Vols. I. to LIV.; of the Memoirs, Vols. L to XXIV., and also Vols. XXVI. to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, XLL, and XLIV. Vols. LV. to LVIIL of the Bulletin, and Vols. XXV., XXX., XXXV, XXXIX., XL., XLIL, XLIIL, XLV. to XLVIII. of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation : — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S. N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905, Lieut. Com- mander L. M. Garrett, U. S. N., Commanding. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. Contributions from the Geological Laboratory, Professor R. A. Daly, in charge. These publications are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology, Cambridge, Mass. ^\^^ Bulletin of the Museum of Compf-ratlve Zoology AT HARVARD CPf^LEG E. Vol. LVJII. N^..i(^' NOTES ON A COLLECTION OF BIRDS FROM YUNNAN. By Outram Bangs and John C Phillips. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. April, 1914. Reports on the SciENt-iFic igseuLTS of the Expedition to the East- ern Tropical P/oific,, in charge op Alexander Agassiz, by the U. S. Fish Commission Si-eamer "Albatross," from October, 1904, to March, 1905, Lmdknant Commander L. M. Garrett, U.S. N,' Commanding, published or in preparation: — A. AGASSIZ. V.' General Report on the Expedition. A. AGASSIZ. I.i Three Letters to Geo. M. Bowers, U. S. Fish Com. A. AGASSIZ and H. L. CLARK. The Echini. H. B. BIGELOW. XVI. i« The Medusae. H. B. BIGELOW. XXIII.23 The Sipho- nophores. H.B. BIGELOW. XXVI.=« The Cteno- phores. B. P. BIGELOW. The Stomatopods. O. CARLGREN. The Actinaria. R. V. CHAMBERLIN. The Annelids. S. P. CLARKE. VIII.8 The Hydroids. . R. COB. The Nemerteans. J. COLE. XIX:i« The Pycnogonlda. . H. DALL. XIV.«< The MoUusks. R. EASTMAN. VII.» The Sharks- Teeth. GARMAN. XII." The Reptiles. . J. HANSEN. The Clrripeds. . J. HANSEN. XXVI1.27 The Schl- zopods. HENSHAW. E. HOYLE. W L. W c. s. H. H. S. W W o. The Insects. The Cephalopods. ~ , C. KENDALL and L. RADCLIPFE. XXV.26 The Pishes. A.KOFOID. III.3 ix.» XX.zo The Protozoa. C. A. KOPOID and J. R. MICHENER. XXII, 2J The Protozoa. C. A. KOPOID and E. J. RIGDEN XXIV." The Protozoa. P. KRUMBACH. The Saglttae. R.VONLENDENPELD. XXI.» The Siliceous Sponges. The Holothurians. The Starfishes. The Ophiurans. G. W. MtJLLER. The Ostracods. JOHN MURRAY and G. V. LEE. XVI 1. 1' The Bottom Specimens. MARYJ. RATHBUN. X.'« The Crus- tacea Decapoda. HARRIET RICHARDSON. Il.« The Isopods. W. E. RITTER. IV.« The Tunlcates. ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. P. E. SCHULZE. XI.u The Xenophyo- phoras. H. R. SIMROTH. The Pteropods and Heteropods. E. C. STARKS. XIII.I3 Atelaxia. TH. STUDER. The Alcyonaria. JH. THIELE. XV.>6 Bathysciadium. T. W. VAUGHAN. VI.« The Corals. R.WOLTERECK. XVIII.n TheAm- phipods. •Bull. M. C. Z., « Bull. M. C. Z., 'Bull. M. C. Z , • Bull. M. C. Z., » Mem. M. C. Z, • Bull. M. O. Z., ' Bull. M. C. Z., « Mem. M. C. Z. • Bull M. C. Z. w Mem. M. C. Z «' Bull. M. C. Z., «2 Bull. M. C. Z., " Bull. M. C. Z., 14 Bull. M. C. Z.. » Bull. M. C. Z.. " Mem. M. C. Z. " Mem. M. C. Z. '» Bull. M. C. Z., w Bull. M. C. Z., >«Bull. M. C. Z., »' Mem. M. C. Z. « Bull. M. C. Z.. » Mem. M. C. Z. 24 Bull. M. C. Z., 26 Mem. M. C. Z !« Bull. M. C. Z.. «' Mem M. C. Z.. Vol. XLVL, No. 4, April, 1905, 22 pp. Vol. XLVL, No. 6, July, 1905, 4 pp., 1 pi. Vol. XLVL, No. 9, September, 1905, 5 pp.. 1 pi. Vol. XLVL, No. 13, January, 1906, 22 pp., 3 pis. , Vol. XXXIII. , January, 1906, 90 pp., 96 pis. Vol. L., No. 3, August, 1906, 14 pp.. 10 pis. Vol. L., No. 4, November, 1906, 26 pp., 4 pis. , Vol. XXXV., No. 1, February, 1907, 20 pp., 15 pis Vol. L., No. 6, February, 1907, 48 pp., 18 pis. , Vol. XXXV, No. 2, August, 1907, 56 pp., 9 pl3. Vol. LI., No. 6. November, 1907, 22 pp., 1 pi. Vol. LIL, No. 1, June, 1908, 14 pp., 1 pi. Vol. LIL, No. 2, July, 1908, 8 pp., 5 pis. Vol. XLIIL, No. 6, October. 1908, 285 pp., 22 pis. Vol. LIL, No. 5, October, 1908, 11 pp., 2 pis. , Vol. XXXVII., February, 1909, 243 pp., 48 pis. , Vol. XXXVIII., No. 1, June, 1909, 172 pp., 5 pis.. 3 maps. Vol. LIL, No. 9, June, 1909, 26 pp., 8 pis. Vol. LIL. No. 11, August, 1909. 10 pp., 3 pla. Vol. LIL. No. 13, September, 1909, 48 pp., 4 pis. , Vol. XLL, August, September, 1910, 323 pp., 56 pla. Vol. LIV., No. 7, August, 1911, 38 pp. . Vol. XXXVIII., No. 2, December, 1911, 232 pp., 32 pis. Vol. LIV., No. 10, February, 1912, 16 pp., 2 pis. , Vol. XXXV., No. 3, April, 1912, 98 pp., 8 pis. Vol. LIV., No. 12, April, 1912, 38 pp., 2 pis. Vol. XXXV, No. 4, July, 1912, 124 pp.. 12 pis. Bulletin of the Museum of Comparative ZoOlogy AT HARVARD COLLEGE. Vol. LVIII. No. 6. NOTES ON A COLLECTION OF BIRDS FROM YUNNAN. By Outram Bangs and John C. Phillips. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. April, 1914. Xo. r>. — Notes on a Collection of Birds from Yunnan. By Outram Bangs and John C. Phillips. The Museum of Comparative Zoology acquired in the autumn of 1912 a series of 1,376 bird skins made by a Japanese collector in south- ern Yunnan. This collection appears to represent well the ornis of the region, and contains, as might be expected, a rather large number of undescribed forms. Mr. Colling\vood Ingram (Xovitates zoologicae, Dec. 1912, 19, p. 269-310) has published a complete list of the birds thus far recorded from Yunnan. The basis of his work was a small collection, "a few hundred specimens," apparently from the same source as our own, the localities and dates being the same. Mr. Ingram's paper mentions from this province 352 species and subspecies, to which we have been able to add seventy-eight, thirteen of which appear not to have been described before. The greater part of our collection was made at Mengtsze, near the southern border of the province, from which the other collecting points, Linan Fu, Shi-ping, and Loukouchai are only a short distance a,way, Mengtsze is an important town, and at the present time the new rail- road runs by within a few miles of it. The town is situated on a plateau of red sand or clay, at an elevation of about 4,500 feet. The plain is some twenty by twelve miles in extent, and is bordered by mountains, at a distance of about a day's journey from the town, which run up to 8,000 feet. IMr. E. H. Wilson, the well-known botanist and traveller, who has visited ^lengtsze, informs us that the country is a rather poor one, the population having been sadly depleted by the ^Slohammetlan war and by bubonic plague. Forested areas are now to be found only on the higher hills, the Mengtsze plain being entirely denuded of trees and composed largely of grass land. The climate is healthy and comparatively cool for the tropics. There is only a short rainy season in mid-summer and the rest of the year is dry and sunny. The region is fairly well watered and there is some artificial irrigation. Rice, maize, sugar-cane, and sweet potatoes are grown, but agricul- turally the country is not at all a rich one. It is probal)le that most of the bird collecting was done in the forest of pine and mixed deciduous trees upon the hills near ^Mengtsze, as 268 bulletin: museum of comparative zoology. Mr. Wilson informs us that the plain about the city is almost birdless. The region appears to be a favorite winter resort for many species of birds, whose breeding grounds are either in the higher mountains near by or in north central and eastern China. The series of a species often contains two subspecies, one, sometimes both of which do not breed in the immediate vicinity. This is made evident by the date on the labels, although no field notes accompanied the collection. Bird col- lecting was carried on through every season of the year, and the com- moner resident species are almost always represented by skins in fresh autumnal or winter dress, and by well-worn summer specimens. We have marked with an asterisk, each species not in Ingram's list. All measurements are given in millimeters, and for all new descrip- tions Ridg\\"ay's color standard (Washington, 1912) has been used. Phasianidae. Francolinus chinensis (Osbeck). 1 cT, Mengtsze, 26 May, 1911. *Arboricola rufogularis euroa, subsp. nov. 2 d^ 's, Mengtsze, March, July. Type:— aduh &, M. C. Z., No. 61,841, Mengtsze, 18 March, 1911. Characters. — General coloration darker and richer, and size larger than in true A. rufogularis Blyth from Sikhim. The white central shaft stripes of the flank feathers are reduced to mere lines; crown greyer and less olivaceous, especially towards the forehead; scapulars and greater wing coverts with larger black areas. Measurements of the type: wing, 160, tarsus, 42, culmen from base of forehead, 25. * Bambusicola oleaginia, sp. nov. Type, and only specimen: — adult cf, M. C. Z. No. 61,839, Mengtsze, 12 December, 1910. Characters. — Somewhat similar to B. fytchii Anderson, of Assam and Upper Burma, but postorbital region black instead of ferruginous ; ground color of upper parts much darker and more olivaceous, with dark central stripes of feathers of the back black instead of ferruginous. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 269 All wing coverts more uniform in color and much darker than in B. fi/ichii; chest more uniform, almost lacking the Avliite spots; black markings of flanks more extensive and more irregular in shape, the feathers of this region often being nearly solid black barred with white; rump and upper tail coverts plain dark olivaceous, wholly lacking the markings of B. ff/tchii. Measurements : wing, 143, tail feathers, 100, culnien, 20.5, tarsus, 48. Gennaeus nycthemerus (Linne). 1 adult 9 , 2 immature, Mengtsze. Phasianus elegans Elliot. 1 adult, cf , Mengtsze, 27 May, 1910. This extends the range of P. elegans in a southwesterly direction. This specimen is like those in the M. C. Z. from western Szechwan taken by the late Walter R. Zappe5\ Treronidae. Sphenocercus sphenurus (Vigors). Adult cT and 9 , IMengtsze, July. The cf agrees fairly well with the two male specimens from northern India in M. C. Z., but the 9 is a much darker green on the upper parts and a slightly richer green below, than in Indian females. COLUMBIDAE. Columba hodgsoni (Vigors). 3 specimens, Loukouchai, Jan., Feb. These agree entirely with a cf from Darjeeling, Indi ia, in M. C. Z. Peristeridae. TuRTUR orientalis Latham. 3 specimens, all immature, Mengtsze, June, July, Sept. 270 bulletin: museum of comparative zoology. Onopopelia humilis (Temminck). 15 specimens, Mengtsze, March, April, May, Sept., Oct., 1910. Spilopelia tigrina (Temminck & Knip) . 6 specimens, Mengtsze, May, July, Aug., Oct. These specimens are more or less intermediate between tigrina and chinensis, as Ingram has pointed out (Nov. zool., 1912, 19, p. 272) in the case of his birds from the same locality, and do not run very true to type. One has very dark shaft stripes on the upper wing coverts but is grey on the under tail coverts. Rallidae. Eulabeornis STRIATU3 JOUYi (Stegnegcr). 1 adult cf , Mengtsze, Aug., 1910. Wing 122; this is maximum for the above race according to Stejne- ger (Proc. U. S. nat. mus., 1886). Limnobaenus fuscus (Linne). 14 specimens, Mengtsze, May, June, 1910. Amaurornis phoenicura (Forster). 2 c^'s, Mengtsze, March, May. Gallinula chloropus orientalis Horsfield. 10 specimens, Mengtsze, Feb., May. Colymbidae. Tachybaptus fluviatilis philippensis (Bonnaterre). 1 adult, 1 immature, Mengtsze, Sept., Nov. bangs and phillips: birds from yinstnan. 271 Laridae. * Hydrochelidon leucopareia swinhoei Mathews. 4 specimens, Mengtsze, 6 June, 1911. These birds are all in immature plumage and probably belong to this race. They are small ; wing 193; 195; 200 and 201 ram. Charadriidae. MiCROSARCOPS CINEREUS (Blyth). 3 specimens, Mengtsze, April, Oct. Pluvialis dominicus fulvus (Gmelin). 2 specimens, cf and 9 , Mengtsze, April, Nov. Charadrius dubius dubius (Scopoli). 4 specimens, Mengtsze, March, Sept. * Charadrius dubius jerdoni (Legge) . 1 adult 9 , Mengtsze, 5 ]March. This specimen has a wing only 101.5 mm. long and therefore would seem to belong to the small race that breeds in the Himalayas. * Trixga tot anus (Linne) . 4 specimens, Mengtsze, 10 Sept., 1910. * Trixga nebularia glottoides (Vigors). 8 specimens, Mengtsze, Sept., Oct., Nov., Dec, 1910. The eastern race of the Green-shanks has been formally recognized by Matthews under the above name. If it can be maintained at all, it certainly is only a very close subspecies. We can find no differences in measurements between eastern and western specimens and no color 272 bulletin: museum of comparative zoology. differences in skins in breeding plumage. In winter plumage eastern skins appear to be just a trifle paler than western ones. * Tringa glareola (Gmelin). 11 specimens, Mengtsze, Aug., Sept., Nov., 1910. Tringa ochropus assami Mathews. 1 9, Mengtsze, 5 Sept., 1910. * Xenus cinereus (Giildenstein). 1 cT, Mengtsze, 19 Sept., 1910. * PisoBiA damacensis (Horsfield). 1 cf, Mengtsze, 25 April, 1911. * LiMOSA LIMOSA MELANUROIDES Gould. 1 cf , Mengtsze, 1 Sept., 1910. We quite agree with Mathews that the eastern Black-tailed godwit is a well-marked race, at once differing from the western form by its much inferior size. Gallinago gallinago gallinago (Linne), 1 adult 9, Mengtsze, 27 Oct., 1910. This specimen is perfectly typical of the western race of the snipe. * Gallinago gallinago uniclavus Hodgson. 4 specimens, both sexes, Mengtsze, Oct., Dec, 1910, all typical of the eastern form (that breeds in northeastern Siberia). * Gallinago strenua (Rulil). 6 specimens, Mengtsze, Sept., Dec. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 273 * SCOLOPAX RUSTICOLA Lmn6. 2 specimens, Mengtsze, March, Dec. ROSTRATULA BENGALENSIS (Linil^). 3 specimens, Mengtsze, May, Sept., Oct. CURSORIIDAE. * Glareola maldivarum (Forster). G specimens, Mengtsze, July, 1910. Gruidae. Megalornis grus lilfordi (Sharpe). 1 adult unsexed, Mengtsze, but without date. Ibididae. *Ibis melanocephalus (Latham). 1 adult 9 , Mengtsze, 1 May, 1910. ^ CiCONIIDAE. * Pseudotantalus leucocephalus (Gmelin). 5 specimens, Mengtsze, Aug., Oct., Nov. Ardeidae. Herodias intermedia Wagler. 6 specimens, Mengtsze, July, Aug. * Nycticorax nycticorax nycticorax (Linnd). 1 cT and 1 9 , IMengtsze, Sept. 274 bulletin: museum of comparative zoology. BUTORIDES JAVANICA JAVANICA (Horsfield). 3 specimens, Mengtsze, Aug., 1910. BuBULCUs coromandus (Boddaert). 2 specimens, Mengtsze, 1 Aug., 1910. IxoBRYCHUs SINENSIS (Gmelin). 1 c? and 1 9, Mengtsze, 16 June, 1911. IxoBRYCHUS cinnamomeus (Gmelin). 7 specimens, Mengtsze, April, May, June. Anatidae. * Nettion crecca (Linn^). 1 9, Mengtsze, Oct. Falconidae. Circus melanoleucus (Forster). 4 specimens, both sexes, Mengtsze, Feb., April, May. * Circus aeruginosus (Linne). 1 specimen, unsexed, Mengtsze, Ma}^ 1910. * Circus spilonotus Kaup. 1 9, Mengtsze, March. AcciPiTER Nisus (Linne). 4 specimens, Mengtsze, June, Oct., Nov. * LoPHOSPiziAS TRiviRGATUS (Tcmminck). 1 cJ", Mengtsze, Sept., 1910. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 275 Cerchneis tinnunculus japonicus (Teminiiick & Schlegel). 1 d' and 1 9 , Mengtsze, 27 Oct., 20 Nov. Cerchneis tinnunculus saturata Blyth. 6 specimens, Mengtsze, 11 March, 8 April, 20 July, 16 Oct., 27 Oct., 25 Nov. These examples are extreme of this large, very dark form, while the two specimens of C. t. japonicus are typical of that race. The birds collected in Szechwan and Hupeh by Mr. W. R. Zappey and referred to saturata by Thayer and Bangs, prove in the light of this material, to be almost exactly intermediate between saturata andjajyonicus. Strigidae. *Otus malayana (Hay). ' 1 o^ and 1 9, adults, Mengtsze, 12 Oct., 16 Oct., 1910. The cf is in the brown phase and the 9 in the grey, d" wing, 85; 9 wing, 93. These birds unquestionably are referable to this species. Ingram recorded Ottis lempiji erythrocamj^e (Swinhoe) from Mengtsze, but our collection from the same place contained no Scops owl belong- ing to that group of the genus. * Nixox scutulata burmanica Hume. 2 adults, 1 9 , one with sex undetermined, Mengtsze, 29 July, 16 Oct., 1910. No Ninox was listed from Yunnan by Ingram, but a specimen from Quaylom, Yunnan, taken by Anderson had been recorded by Sharpe (Cat. birds Brit, mus., 2, p. 162). Our two skins appear to belong to this form, which even Hartert (Vogel Palaark. fauna) says he does not know very well. They are much larger than two Malacca trade skins of A^. scutulata malaccensis (Eyton) in M. C. Z. Compared with several specimens of N. scutu- lata scutulata (Raffles) from the Riu Kiu Islands and the Philippines the Yunnan birds while of the same size, are of a different shade of brown above and have very greyish heads. They are also somewhat 276 bulletin: museum of comparative zoology. different below, the ground is slightly paler and the markings, es- pecially on the belly, have a more transverse and less longitudinal appearance. CORACIIDAE. * EuRYSTOMUS orientalis calonyx Sharpe. 1 d", Mengtsze, 12 Oct., 1910. Alcedinidae. Alcedo ispida bengalensis Gmelin. 8 specimens, Mengtsze, Feb., April, June, Aug., Nov. * Halcyon pileatus (Boddaert). 5 specimens, Mengtsze, April, Sept., Oct. Halcyon smyrnensis fuscus (Boddaert). 24 specimens, Mengtsze, May, June, July, Aug., Sept., Oct., Nov. Meropidae. Merops philippinus Linne. 13 specimens, Mengtsze, April, July, Aug., Sept. Upupidae. Upupa epops saturata Lonnberg. 10 specimens, Mengtsze, Feb., March, Aug., Sept., Oct. All of these are referable to this form which is probably only a migrant in southern Yunnan. Upupa epops, sub. sp.? 14 specimens, Mengtsze, May, June, July, Aug., Oct., Nov. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 277 This series represents the breeding Hoopoe of the region. While nearer to saturata than to any of the other subspecies, this race cannot quite be referred to that form. The specimens in the series do not run true. Some approach indica in the character of the crest-feathers, although not nearly so reddish in general coloration as are extreme examples of that form. Others can hardly be separated from Euro- pean examples of true epops. While others again might easily pass for saturata. jMany of the specimens were in worn breeding plumage when killed. We regret leaving this bird without a name, and perhaps all speci- mens from the region should be referred to saturata, with the state- ment that the breeding individuals are not typical, but such a course seems unscientific. Caprimulgidae. Caprimulgus monticola Franklin. 1 cf , Mengtsze, 1 Aug., 1910. * Caprimulgus macrurus ambiguus Hartert. 1 9, Mengtsze, 4 Dec, 1910. Micropodidae. Apus affinis subfurcatus (Blyth). 7 specimens, Mengtsze, April, June; Loukouchai, June. CUCULIDAE. * HiEROCOCCYX SPARVEROIDES VigOrS. 1 specimen, without data. * CucuLUS CANORUS TELEPHONUS Heine. 3 specimens, Mengtsze, 3 to 15 May, 1911. * CucuLUS CANORUS BAKERi Hartert. 6 specimens, jVIengtsze, April, May, June, July. 278 bulletin: museum of comparatR'E zoology. These skins agree exactly with Hartert's description, showing all the characters of the subspecies. This form is the breeding bird of the region, and ielephonus, of course, is only a migrant in southern Yunnan. CUCULUS OPTATUS Gould. I adult d^, Mengtsze, 8 April 1911. Cacomantis merulinus (Scopoli). II specimens, Mengtsze, May, Aug., Sept. Eudynamis orientalis honoratus (Linne). 17 specimens, Mengtsze, April, INIay, June, July, Sept., Oct. Capitonidae. Cyanops davisoni (Hume). 1 cf 1 9 , Loukouchai, Feb., June. * CyanoPvS franklini (Blyth). 6 specimens, Loukouchai, Jan., Feb., 1911. PiCIDAE. Picus CANUS soRmDiOR Rippon. 8 specimens, Mengtsze, March, July, Nov.; Shi-ping, Feb., March. Dryobates hyperythrus subrufinus (Cabanis & Heine), 1 9 , ISIengtsze, 14 March. The present example of this rare bird agrees exactly with the two specimens from Szechwan collected by Mr. W. R. Zappey. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 279 Dryobates cabanisi cabanisi (Malherbe). 5 specimens, Mengtsze, Oct., 1910, Jan., 1911; Shi-ping, Feb., March, 1911; Linan Fu, Feb., 1911. Dryobates pygmaeus scintilliceps (Swinhoe). 10 specimens, Linan Fu, Feb.; Mengtsze, March, April, Sept., Nov.; Loukouchai, Feb. Picumnus innominatus chinensis (Hargitt). 3 cf's, Mengtsze, April, 1911; Loukouchai, Dec, 1910, Feb., 1911. These skins are slightly smaller and in color somewhat brighter on the lower back than a specimen collected by Mr. W. R. Zappey in Hupeh. * Sasia ochracea Hodgson. 1 adult cf, Loukouchai, Jan., 1911. Jynx torquilla japonica Bonaparte. 10 specimens, Mengtsze, Oct., Nov.; Loukouchai, Dec; Shi-ping, Feb. Hirundinidae. Chelidon rustica gutturalis (Scopoli). 8 specimens, Mengtsze, April, May, June, July, Oct.; Loukouchai, Dec Chelidon tytleri (Jerdon). 4 specimens, Mengtsze, Dec, 1910. * Chelidon daurica striolata (Temminck & Schlegel). 3 specimens, Mengtsze, June, 1911. 280 bulletin: museum of comparative zoology. muscicapidae. * Cyornis tickelliae glaucicomans Thayer & Bangs. 6 specimens, Mengtsze, April, June, Aug., Sept., Oct. * NiLTAVA SUNDARA DENOTATA, Subsp. nOV. 2 specimens, cf and 9 , Mengtsze, Oct., Dec. Tijpe:— adult &, M. C. Z., No. 61,905, Mengtsze, 14 October, 1910. Characters. — Similar to true A^ sundara but slightly larger; back pure black, not washed with purple as in the latter form; underparts paler yellowish especially on the lower abdomen; chin and throat solid black, not at all washed with purple; neck spot pale blue. (Rood's blue, Ridgway, 1912); blue of head abruptly contrasted with the black of the mantle. Measurements: cT, Tyjjc, wing, 83, tarsus, 21, culmen from base of forehead, 16; 9 , No. 61,906 M. C. Z., wing, 78, tarsus, 22, culmen, 17. * NiLTAVA DAViDi La Touchc. 3 specimens, both sexes, Mengtsze, April, Oct. Thayer and Bangs (Bull. M. C. Z., 1909, 52, p. 141) overlooked La Touche's name when describing their Niltava lychnis. There is, however, a chance that their bird from Hupeh is subspecifically dis- tinct from N. davidi La Touche, which came from northwestern Fokien. The two males in the present collection agree almost exactly with La Touche's d'escription of N. davidi and differ from the type of N. lychnis in having the crown brighter blue and the chin, sides of the head and throat washed with dark, bluish purple, whereas in the type of N. ly chilis these areas are dead black. MusciCAPA LATiROSTRis (Raffles). 5 specimens, Mengtsze, Feb., May, Sept. * MusciCAPA muttui Layard. 1 d", Mengtsze, 14 April, 1911. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 281 MUSCICAPA STROPHIATA HodgSOn. 2 specimens, 1 cf and 1 9, Mengtsze, Nov., 1910, April, 1911. * MUSCICAPA PARVA ALBICILLA Pallas. 4 specimens, Mengtsze, April, Oct. INIUSCICAPA MELANOLEUCA MEL.VNOLEUCA HodgSOn. 3 specimens, Mengtsze, March, Oct. * MusciCAPA MUGiMAKi Teniminck. I d", Mengtsze, April, 1911. Chelidorhynx hypoxantha (Blytli). 4 specimens, ^Mengtsze, Feb., March, Dec. Rhipidura albicollis (Vieillot). II specimens, ^Mengtsze, Feb., March, April, July; Loukouchai, Jan. * Hypothyimis azurea azurea (Boddaert). 1 cf and 1 9 , Mengtsze, 6 Sept., 16 Oct. * Cyanoptila cyanomelana (Temminck). 1 cf, immature, Mengtsze, 18 Oct., 1911. TCHITREA INCH (Gould). 11 specimens, Mengtsze, April, Aug., Sept., Oct. CuLiCAPA CEYLONENSis Swainson. 2 specimens, Mengtsze, Oct. 282 bulletin: museum of comparative zoology. Crypto LOPHA burkii tephrocephalus (Anderson). 7 specimens, Mengtsze, March, April, Oct. * Cryptolopha trivirgatus eiuncidus, subsp. nov. Type, and only specimen: — adult d^, M. C. Z., No. 61,985, Mengtsze, 16 September, 1910. Characters. — Similar to true C. trivirgatus (Strickland) ; but general color of underparts very much paler and brighter yellow, clear lemon-, yellow on throat and chin; sides of head clearer and paler yellow; the black stripes on the top of the head narrower anteriorly; general color of upper parts more yellowish and less greenish; rump and upper tail coverts especially so. Measurements: wing, 55, tarsus, 17, exposed culmen, 9. Stoparola melanops (Vigors). 13 specimens, Mengtsze, March, April, July, Aug., Sept., Oct.; Loukouchai, June. Campophagidae. Pericrocotus brevirostris ethologus, subsp. nov. 17 specimens, both sexes, Mengtsze, Feb., March, April, July, Sept.; Shi-ping Feb.; Loukouchai, Dec. On comparison of a large series of specimens it seems best to divide P. brevirostris (Vigors) into three subspecies, based on the coloration of the adult male plumage. The forms are about alike in size, and the females of the three are rather similar. Oates (Fauna of British India), has already alluded to the differences in color shown by the two Indian forms. The Chinese is quite different from either and has the under wing coverts and axillars nearly as yellow as in Pericrocotus igneus Blyth although otherwise not in the least like that species. Vigors's description and Gould's figure, both based on the same specimens, clearly were taken from the eastern form of intense colora- tion; M'Clelland's P. affinis is synonymous. The Yunnan skins are not typical of the Chinese form as defined below, and though nearer to it than to true P. brevirostris are inter- BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 283 mediate, as might be expected on geographical grounds, between the two. The three races may be briefly defined as follows : — 1. Pericrocoius brevirostris brevirostris (Vigors). Range. — Eastern Himalayas from Sikkim and Assam, through Manipur etc. to Tenasserim. Characters. — Adult cf with the red of underparts and rump intense scarlet, sometimes almost scarlet-red; under wing coverts and axil- lars Grenadine red. 2. Pcrierocotus brevirostris flavillaceus, subsp. nov. Type: — Adult &, M. C. Z. No. 24,146, Koolloo Valley, northern India. M. M. Carleton. Range. — More western Himalayas and Plains of India. Characters. — Adult c? with the red of underparts and rump, in- cluding the under wing coverts and axillars Grenadine red. 3. Pericrocoius brevirostris ethologus, subsp. nov. Ty2je:—Adu\t &, M. C. Z., No. 51,487, Hsienshan, Hupeh, China, May 28, 1907, W. R. Zappey. Range. — Central, western, and northern China. (Birds from south- ern Yunnan are not typical). Characters. — Adult d^ with the red of underparts and rump, flame scarlet, much mixed with orange on chest, and with white on belly; under wing coverts and axillars much yellower than in the other forms — orange-chrome; throat decidedly more greyish black than in the other races. Pericrocotus roseus (Vieillot). 2 cf 's, Mengtsze, March, Oct. * Pericrocotus cantonensis Swinhoe. 6 specimens, Mengtsze, April, Oct. * Campophaga lugubris (Sundeval). 3 specimens, Mengtsze, March, Oct. Campophaga melanoptera (Riippell). 8 specimens, INIengtsze, March, April, May, Oct. 284 bulletin: museum of comparative zoology. Pycnonotidae. Chloropsis hardwickei Jardin & Selby. 5 specimens, Loukouchai, Jan., Feb. Hypsipetes leucocephalus (Gmelin). 16 specimens, Mengtsze, March, April, Nov. loLE HOLTi Swirihoe. 16 specimens, Mengtsze, Jan., March, April; Loukouchai, Jan., Feb. Criniger tephrogenys henrici Oustalet. 5 specimens, Loukouchai, Feb. * Alcurus striatus paulus, subsp. nov. 2 specimens, Loukouchai, Feb. Type: — Adult d', M. C. Z. No. 62,006 Loukouchai, 5 February, 1911. Characters. — Similar in color to true striatus of the Himalayas, but size much less. Measurements : - wing culmen tarsus tail featliers cf" 62,006 102 15 18 90 9 62,007 93 13 18 83 Remarks. — Gates (Fauna of British India, 1889) pointed out the difference in size between birds from the Himalayas and those from Tenasserim and Manipur. Our specimens agree in measurements with those from the latter region and are so much smaller than Hima- layan examples as to leave no choice but to give this little form a name. Xanthixus flavescens (Blyth). 1 cf, Loukouchai, 7 Feb., 1911. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 285 Pyconotus atricapillus (Vieillot). 2 specimens, Mengtsze, March, April. Pyconotus xanthorrhous J. Anderson. 5 specimens, Mengtsze, Jan., 1911 ; Loukouchai, Feb. * Spizixus semitorques Swinhoe, 5 specimens, Loukouchai, Jan., Feb., Dec. Spizixus canifrons ingrami, subsp. nov. 6 specimens, Mengtsze, March, Aug., Sept.; Loukouchai, Dec. Type: — Adult d", M. C. Z. No. 62,008, Mengtsze, 18 March, 1911. Characters. — Similar to true canifrons of the Khasi Hills, Manipur; but throat grey and not brown; ear coverts pale ashy grey; under- parts dull olive-green, not greenish yellow as in true canifrons. Size similar. M< jasurei nents : wing culmen tarsiis tail feathers c^ 62,008 95 12 20 92 9 62,009 topotype 92 11 19 82 Remarks. — All the above characters were noticed by Ingram but he hesitated to give the Yunnan form a name on account of insufficient material. TiMELIIDAE. L\NTHOCiNCLA CANORA (Linne). 1 d", Loukouchai, 12 June, 1911. L\NTHOCINCLA CINEREICEPS STYANI Oustalct. 1 cf , Loukouchai, 6 Feb. * L\.NTHOCINCLA LUSTRABILA, Sp. nOV. Type, and only specimen: — Adult c^, M. C. Z. No. 62,014, Loukou- chai, 11 February, 1911. 286 bulletin: museum of comparative zoology. Characters. — Somewhat like I. viilni David of Fokien but ear coverts grey instead of white; Avhole top of head and hind neck bril- liant mars-orange (Ridgway, 1912) forming a cap sharply contrasted with the olive-colored mantle; lower back, rump, and upper tail coverts medal bronze instead of "golden olive" (David). Measurements: wing, 99, culmen, 17, tarsus, 40, tail feathers, 107. POMATORHINUS MACCLELLANDI ODICUS, Subsp. UOV. 7 specimens, Shi-ping, Feb., June, Aug., Sept., Oct. Type: — d^ adult, Mengtsze, M. C. Z., No. 61,999, 22 June, 1911. Characters. — Differs from both true P. m. macclellandi Jerdon from As; am and P. ?«. gravivox David from central China in having sides, flanks, and under tail coverts uniform rich orange rufous. Throat and foreneck unspotted as in true macclellandi, thus differing from gravivox. Measurements : wing culmen tarsus tail feathers d' 61,999 84 28 35 93 & 62,000 90 30 35 90 & 62,001 87 . 29 36 95 & 62,002 86 28 34 96 9 62,003 87 27 37 90 9 62,004 89 29 36 95 & 62,005 90 30 34 95 POMATORHINUS RUFICOLLIS RECONDITUS, Subsp. nov. 9 specimens, Mengtsze, Mar., Sept., Nov.; Shi-ping, March; Lou- kouchai, Feb. Type: — Sid\x\t d^, M. C. Z. No. 62,046, Mengtsze, 22 November, 1910. • Characters. — Similar to true P. ruficoUis Hodgson of the Hima- layas and to P. ruficoUis styani Seebohm of the Yangtsze Valley, but differ ng from both in the co'or of the bill which is wholly yellow except for a small dusky patch at base of upper mandible; in color differing from true ruficoUis in being much more olivaceous and less reddish brown above, tail and wings more olivaceous and less reddish brown; stripes on breast much more distinct and in the adult bright ferruginous; from styni the new form differs in color in being very BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 287 slightly darker on the upper parts the striping below is very different, being more distinct and much more ferruginous. Our bird differs from P. stridulus Swinhoe which Hartert regards as a species, in lack- ing the sharp contrast between the colors of the back and tail and in having a much longer bill. Measurements : culmen wing to base of forehead tarsus tail feathers & 62,046 74 23 30 85 cf 62,047 76 23 30 85 cf 62,048 80 24 30 83 d" 62,054 72 24 29 80 Dryonastes saxnio (Swinhoe). 23 specimens, Mengtsze, Jan., April, May, June, July, Aug., Sept.; Loukouchai, Jan., Jime. The specimens of this long series differ a little in color from those taken by Zappey in Hupeh and Szechwan, being slightly more oliva- ceous and less rusty brown on the upper side. We have seen no specimens of D. sannio from the type locality, Amoy, if it and the Hupeh birds are the same, then D. albosuperciliaris of Godwin and Austen from IManipur Valley near Kaibi, which is usually thrown into sjTionymy, may apply to our Yunnan birds. Pyctorhis sinensis (Gmelin). 22 specimens, Mengtsze, Jan., April, May, June, Sept., Nov., Dec; Loukouchai, June, Dec. * Alcippe nepalensis hueti (David). 11 specimens, Mengtsze, Jan., Feb. These birds cannot be distinguished from specimens of true hueii taken in Hupeh and Szechwan by Zappey. Proparus genestieri (Oustalet). 13 specimens, Mengtsze, Jan., ]\Iarch, April, May, Aug., Nov.; Loukouchai, Feb. 288 bulletin: mu;seum of comparative zoology. The average wing measurement of thirteen specimens is only 58 mm. The original description of this bird (Bull. Mus. Paris, 1897, 3, p. 210), gives the wing at 70. Ingram says that four of his five specimens are small, but does not give measurements. Stachyris nigriceps Hodgson. 1 d', Loukouchai, 10 Feb., 1911. Stachyrhidopsis ruficeps (Blyth). 4 specimens, Mengtsze, Feb., Dec. MixoRNis RUBRiCAPiLLA (Tickell). 1 cf , Mengtsze, 18 June, 1911. * Myiophoneus caeruleus (Scopoli). 1 adult cf , Mengtsze, 3 May, 1911. Myiophoneus eugenei Hume. 8 specimens, Mengtsze, March, 1911; Loukouchai, Feb., Dec. * Brachypteryx cruralis senensis Rickett. 1 adult c^, Mengtsze, 28 ^March, 1911. This example agrees exactly with Rickett's description of the north- western Fokien bird. It probably is only a migrant in Yunnan. * ACTINODURA RAMSAYI YUNNANENSIS, subsp. nov. 18 specimens, Mengtsze, May; Loukouchai, Jan., Feb., Dec. Ty2)e: — adult d', M. C. Z. No. 62,025, Loukouchai, 29 January, 1911. Characters. — Similar to true .1. ramsayi Walden of the Karen-nee Hills, Burma, but much smaller, wing 81 to 87, tarsus 24 to 30, and color of crown much deeper brown (Sudan brown to antique brown, Ridgway, 1912). BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 289 Measurements : wing culmen tarsus tail feathers cf 02,025 85 14 30 112 9 62,015 85 14 28 114 Malacias desgodixsi (David & Oustalet). 4 specimens, Loukouchai, Feb. Siva cyanuroptera wingatei Ogilvie-Grant. 19 specimens, Mengtsze, March, April, Aug., Sept., Oct., Nov., Dec. In this long series the character given by Ingram is very apparent, viz. : — " Secondaries never margined posteriorly, with a distinct white edge as in cyamiropiera." They are however sometimes slightly tipped with white. Siva castaneicauda Hume. 1 cf and 1 9 , Mengtsze, Jan., Feb. YuHiNA DiADEMATA Vcrreaux. 4 specimens, Loukouchai, Feb., Dec; Mengtsze, March. YuHiNA OCCIPITALIS Hodgson. 2 specimens, Mengtsze, Jan. * IXULUS FLAVICOLLIS ROUXI Oustalct. 10 specimens, Mengtsze, Jan., Feb., March. Leiothrix lutea lutea (Scopoli). 3 specimens, Mengtsze, Feb. Pterythius aeralatus Tickell. 1 cf , Loukouchai, 13 Feb., 1911. 290 bulletin: museum of comparative zoology. * Pterythius melanotis Hodgson. 1 & and 1 9 , Loukouchai, Feb. Mesia argentauris Hodgson. 1 c?, Loukouchai, 6 Feb. * MiNLA jERDONi J. Verreaux. 8 specimens, adults of both sexes, Mengtsze, Jan., Feb., Sept. Ingram refers the specimens examined by himself, from western Yunnan to Minla igneitinda Hodgson. Our skins, however, all winter and autumn killed, belong to the Chinese form having olive backs, and wholly bright yellow underparts and under wing coverts, the sides flecked with olive-green. Probably the species is a winter visitor only to our region. * Paradoxornis guttaticollis David. 3 males, Mengtsze; Loukouchai, Feb., March, Dec. * Suthora webbiana webbiana Graj'. 3 specimens, both sexes, Loukouchai, Jan., Feb., 191 L On geographical grounds one might expect these skins to be refer- able to S. w. suffusa Swinhoe. Such, however, is clearly not the case. The crown and edges of the primaries in all three examples is rich vinaceous rufous and the back is of the shade peculiar to true webbiana and different from the shade shown in siiffusa. All three were taken in winter and the collection contains no summer nor spring examples; it is therefore, we think, safe to assume that they were merely winter stragglers of the northern form. The specimens afford the following measurements in mm. wing ciilmen tarsus tail feathers cf ad. 62,956 49. 8 19.5 55. 9 ad. 62,957 50. 8.5 20. 60. 9 ad. 62,958 47. 7.5 19. 58. bangs and phillips: birds from yuxxan. 291 Troglodytidae. * Pnoepyga pusilla Hodgson. I adult 9, Mengtsze, 18 March, 1911. We have no specimens for comparison. The adult 9 seems rather larger than measurements given for Indian birds and it may prove to belong to a different race. Wing, 50, tarsus, 19, exposed culmen, 9. TURDIDAE. Oreocincla aureus (Holandre). 5 specimens, Mengtsze, Jan., Feb., Oct. * TuRDUS OBSCURUS Gmclin. 6 specimens, Mengtsze, Oct., Xov. TuRDUS FUSCATUS Pallas. 12 specimens, Mengtsze, INIarch, Nov.; Shi-ping, March; Linan Fu, Feb., Loukouchai, Dec. TuRDUS merula maxdarixus Bonaparte. II specimens, Shi-ping, March; Linan Fu, Feb. Monticola erythrogaster (Vigors). 1 adult cf, Loukouchai, 8 Dec, 1910. Moxticola solitarius paxdoo (Sykes). 19 specimens, Mengtsze, Jan., March, July, Nov.; Loukouchai, Jan., Feb., June, Dec; Shi-ping, Feb.; Linan Fu, Feb. The series taken by ]Mr. Zappey in Hupeh and Szechwan should have been referred to this race (Mem. M. C. Z., 40, p. 175). 292 bulletin: musel^i of comparative zoology. * Enicurus guttatus bacatus, subsp. nov. 1 cf and 1 9 , Loukouchai, Feb. Type:— \du\t 9, M. C. Z. Xo. 62,033, taken at Loukouchai, 14 February, 1911. Characters. — Very similar to true giittafus Gould, of India; size about the same, but immediately recognized by the much larger and more numerous white spots on the back, the maximum diameter of which is 4 mm. * Enicurus schistaceus Hodgson. 7 specimens, Loukouchai, Feb., Dec. Chaimarrornis leucocepil^la (Vigors). 14 specimens, Mengtsze, Nov., Jan., Feb., March; Loukouchai, Dec, Jan., Feb., April. Rhyacomis fuligixosa fuligixosa (Vigors). 14 specimens, Mengtsze, Dec, Jan., Feb., Nov.; Loukouchai, Feb. Phoexicurus auroreus leucopterus (Blyth). 18 specimens, Mengtsze, Jan., March, Oct., Nov., Dec. Phoexicurus froxtalis Vigors, 2 specimens, Mengtsze, March. Ianthia cyaxura (Pallas). 11 specimens, Loukouchai, Feb., Dec; Mengtsze, Jan., April, Nov., Dec. * Ianthia practica, sp. nov. 1 cf and 1 9 adults, Mengtsze, April; Loukouchai, Feb. Type:— No. 62,035 M. C. Z., adult d", Loukouchai, 14 February, 1911. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 293 Characters. — Adult male most like the adult male of /. rufulata Hodgson but the color of the whole upper parts is dark tyrian blue instead of dark ultramarine blue (Ridgway 1912). The superciliary stripe is shining Chapman's blue instead of Rood's blue; shoulder patch somewhat brighter blue; base of the feathers of superciliary region dusky instead of white; white areas of lower parts much whiter and not so clouded with dusky. Size similar except the bill, which is smaller and more slender. irements : wing cnlmen tarsus tail feathers cf ■ 62,035 type 83 9 25 63 9 62,036 82 9 25 65 "' Calliope calliope (Pallas) . 11 specimens, Mengtsze, April, May. NOTODELA LEUCURA (HodgSOn) . 2 cf's, Mengtsze, July, Aug. COPSYCHUS SATJLARIS SAULARIS (LinU^) . 20 specimens, Mengtsze, Feb., March, April, Oct.; Loukouchai, Dec; Linan Fu, F'eb.; Shi-ping, March. Saxicola torquata ixdica (Bl^'th). 8 specimens, Mengtsze, Aug., Nov., Dec; Loukouchai, Feb. Ingram referred his one male from ISIengtsze to this form. Our skins probably also belong here, although they seem somewhat to approach S. torquata stcjnegeri Parrot and may be intermediate. All being in winter or in immature plumage renders positive identification rather difficult. They are small, the wing in the series of eight skins, ranging from 67 to 70 mm. Saxicola caprata bicolor (Sykes). 2 cf's, Mengtsze, Feb., March. 294 bulletin: museumof comparative zoology. Oreicola ferrea haringtoni Hartert. 11 specimens, Mengtsze, Jan., Feb., March, Oct., Nov., Dec. Sylviidae. Sutoria sutoria phyllorrhaphea (Swinhoe). 8 specimens, Mengtsze, April, Maj^ July, Aug. * AcROCEPHALUS ARUNDINACEUS ORiENTALis Temminck & Schlegel. 2 (f's, Mengtsze, April, Aug. * CiSTicoLA cisticola tintinnabulans (Swinhoe). 10 specimens, Mengtsze, Jan., March, April, May, July. Franklinia gracilis (Franklin). 8 specimens, Mengtsze, Jan., April, Maj^, Sept. Oreopneuste subaffinis (Ogil vie -Grant). 1 cf , Mengtsze, 3 July, 1910. Oreopneuste fuscatus (Blyth). 5 specimens, Mengtsze, April, May. Reguloides proregulus proregulus (Pallas). 1 c/^ and 1 9, Mengtsze, June; Loukouchai, Dec, 1910. Reguloides superciliosus superciliosus (Gmelin). 4 specimens, Mengtsze, April, July, Sept., Oct. Acanthopneuste borealis borealis (Blasius). 1 9, March, 1911. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 295 ACANTHOPNEUSTE LUGUBRIS (Blyth). 2 cf 's, Mengtsze, April, May. * AcANTHOPNEUSTE CORONATA (Temininck) . 4 specimens, Mengtsze, Aug. AcANTHOPNEUSTE TROCHILOIDES (Sundeval). 1 cf and 1 9 , Mengtsze, June, Oct. AcANTHOPNEUSTE DAVISONI Oates. 1 cT, Mengtsze, 21 Oct., 1910. Phyllergates coronatus (Jerdon & Blyth). 1 9 , Mengtsze, 29 July, 1911. * HoRORNis canturians Swinlioe. 1 d', Mengtsze, 18 Nov., 1910. * HoRORNis FORTiPES DAViDiANA (Verreaux) . 2 cf 's, Mengtsze, May, June. SuYA suPERCiLiARis Andcrson. 7 specimens, Mengtsze, April, May, Aug. SuYA CRiNiGERA YUNNANENSis Harrington. 3 specimens, Mengtsze, March, May. These skins are very dark in color, and agree with the account of this recently described form. Prinia inornata exter Thayer & Bangs. 12 specimens, Mengtsze, Jan., March, April, June, Sept. This series agrees with specimens from Szechwan collected by Zap- pey, although the Yunnan specimens average a trifle smaller and paler. 296 bulletin: museum of compakative zoology Laniidae. Lanius hypoleucus Blyth. 3 specimens, Mengtsze, 19, 23, 25 Aug. * Lanius schach schach (Linne). 14 specimens, Shi-ping, 4 March; Loukouchai, 20, 21, 22 Dec; Mengtsze, Jan., March, June, July, Aug., Sept. Lanius schach tephronotus (Vigors). 1 cf , Loukouchai, 21 Dec. Lanius tigrinus Drapiez. 3 specimens, Mengtsze, April, Aug. Lanius nasutus nigriceps Franklin. 1 d", Linan Fu, 20 Feb. * Lanius fuscatus Lesson. 1 specimen, without exact data. Lanius cristatus cristatus Linne. 4 specimens, Mengtsze, May, Sept., Oct.; Loukouchai, Dec. * Lanius cristatus superciliosus Latham. 1 9 , Mengtsze, 12 May. Paridae. Parus rex David. 4 specimens, Loukouchai, Feb., March. Parus major commixtus Swinhoe. 14 specimens, Mengtsze, March, April, Aug., Sept., -Oct., Nov., Dec; Loukouchai, Dec; Linan Fu, Feb. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 297 Aegithaliscus concinnus Gould. 9 specimens, Mengtsze, Jan., Feb., March, Aug., Dec. SiTTIDAE. SiTTA EUROPAEA MONTIUM La Touche. 1 cf , Loukouchai, 25 Jan. Dendrophila frontalis (Horsfield). 4 specimens, Loukouchai, Feb., June, July. ZOSTEROPIDAE. ZosTEROPS palpebrosa mussoti Oustalet. 10 specimens, Mengtsze, Jan., April, May, July, Aug., Sept. The wing averages 52 in this series, just as in Oustalet's series from Szechwan, and the color agrees perfectly with his description. Ingram has suggested that his Yunnan birds may be referable to this race and there seems to us no doubt that this is so. DiCAEIDAE. DiCAEUM IGNIPECTUS IGNIPECTUS (HodgSOn). 4 specimens, Mengtsze, Jan., March; Loukouchai, Feb. * DiCAEUM OLiVACEUM Waldeu. 7 specimens, Mengtsze; Loukouchai, April, June, Sept., Oct. Nectariniidae. Aethopyga sanguinipectus Walden. 13 specimens, Loukouchai, Feb.; Asanzai, April. Aethopyga dabryi Verreaux. 28 specimens, Mengtsze, Feb., March, April, Aug. 298 bulletin: museum of comparative zoology. * Arachnothera magna (Hodgson). 4 specimens, Loukouehai, Feb. Motacillidae. Motacilla alba hodgsoni Blyth. 3 specimens, Mengtsze, Sept., Oct. * Motacilla alba ocularis Swinhoe. 7 specimens, Mengtsze, Feb., April, June. Motacilla boarula melanope Pallas. 6 specimens, Mengtsze, April, May, Sept., Oct., Nov. Budytes cttreola citreola (Pallas). 4 specimens, Mengtsze, April. * Budytes citreola citreoloides Gould. 1 cf, Mengtsze, 7 March, 1911. Dendronanthus indicus (Gmelin). 3 specimens, Mengtsze, May. Anthus trivialis hodgsoni Richmond. 5 i-pecimens, Mengtsze, Jan., April, Oct., Nov. Anthus richardi richardi Vieillot. 3 specimens, INIengtsze, March, Oct. Anthus rufulus rufulus Vieillot. 5 specimens, Mengtsze, March, April, Aug., Oct. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 299 Anthus roseatus Blyth. 1 d^ and 1 9, Mengtsze, April. * Oreocorys sylvanus (Hodgson). I adult cf, Mengtsze, 5 June, 1911. Alaudidae. Alauda gulgula coelivox Swinhoe. 3 specimens, Mengtsze, April, Oct. Wing, 89 to 92. These specimens are strongly and darkly marked on the upper breast and are in general color very reddish. They probably belong to this race, although our series is small for compari- son. Frixgillidae. * EOPHONA MELANURA MIGRATORIA Hartcrt. 5 specimens, Linan Fu, Feb. Spinus ambiguus (Oustalet). 10 specimens, Mengtsze, Jan., Feb., March, Dec. Passer rutilans cinnamomea (Gould). 3 specimens, Mengtsze, April; Linan Fu, Feb. Passer montanus montanus Linne. II specimens, Mengtsze, April, Aug., Oct., Nov. Carpodacus erythrinus roseatus (Hodgson). 7 specimens, Mengtsze, Feb., March, April, Dec. * LOXIA CURVIROSTRA HIMALAYENSIS (Blyth). 1 adult c?, Mengtsze, 20 March, 1911. 300 bulletin: museum op comparative zoology. Emberiza pusilla Pallas. 10 specimens, Mengtsze, Jan., March, April, Nov., Dec. Emberiza fucata arcuata Sharpe. 2 specimens, Mengtsze, March, April. Emberiza aureola Pallas. 3 specimens, Mengtsze, April, May. * Emberiza rutila Pallas. 3 specimens, Mengtsze, April; Loukouchai, Jan. Emberiza spodocephala melanops Blyth. 5 specimens, Mengtsze, Jan., April, Oct., Dec. Melophus melanicterus (Gmelin). 25 specimens, Mengtsze, March, April, May, Aug.; Loukouchai, Dec; Linan Fu, Feb.; Shi-ping, Feb. Ploceidae. MuNiA punctata topela Swinhoe. 14 specimens, Mengtsze, Jan., Feb., March, April, Aug., Nov.; Loukouchai, Dec. Sporaeginthus flavidiventris (Wallace). 10 specimens, Mengtsze, March, May, June, July; Loukouchai, June. Sturnidae. Spodiopsar nemoricolus (Jerdon). 12 specimens, Mengtsze, April, June, Aug., Oct. BANGS AND PHILLIPS: BIRDS FROM YUNNAN. 301 * Spodiopsar sericeus (Gmelin). 1 d^, LinanFu, 20 Feb., 1911. Aethiopsar cristatellus Gmelin. 23 specimens, Mengtsze, Jan., March, Aug., Sept., Dec. Oriolidae. Oriolus indicus Jerdon. 7 specimens, Mengtsze, April, May, Sept., Oct. Dicruridae. Chibia hottentotta (Linne). 3 specimens, Mengtsze, Sept., Oct. BUCHANGA ATRA CATHOECA Swinhoe. 21 specimens, Mengtsze, April, IMay, June, July, Aug., Oct. BucHANGA cineracea pyrrhops (Hodgson). 12 specimens, Mengtsze, Jan., Feb., March, April, Oct.; Shi-ping, Feb., March. These specimens are similar in color to true cineracea but are larger than specimens of that form from Java. They agree with measure- ments given by Sharpe (Cat. birds Brit, mus.) for jW^^oj^s of Hodgson, by which name they undoubtedly should be known. * BucHANGA leucogenys leucogenys Waldcu. 4 specimens, Mengtsze, Oct. This series agrees with birds from Cochin China and the Malay Peninsula in being dark grey in color. It has been suggested by Hartert (Nov. zool., 17, p. 248) that there may be a northern and southern race. 302 bulletin: museum of comparative zoology. Our material now shows that such is the ease, and that the series of birds collected by Mr. W. R. Zappey in Szechwan and Hupeh are very much paler, almost whitish grey in color. For this pale northern form we propose the name Buchanga leucogenys cerussata, subsp. nov. Type: — Adult cf , M. C. Z. 52,035 Ichang, Hupeh, China, June 19, 1907, W. R. Zappey. Characters. — Similar in size to B. I. leucogenys, but much paler and more whitish grey throughout, with light face area larger and more clearly defined. CORVIDAE. CoLOEus DAUURicus (Pallas). 3 specimens, Mengtsze, June, Dec. CoLOEus NEGLECTUS (Schlegel). 1 unsexed, Mengtsze. Pica pica sericea Gould. 10 specimens, Mengtsze, Jan., Feb., March, May, June, Dec. Urocissa erythrorhyncha (Gmelin). 10 specimens, Mengtsze, Feb., March, April, Oct., Dec. DendrociTta himalayensis (Vigors). 2 specimens, Loukouchai, June. The following Publications of the Museum of Comparative Zoology are in preparation: — • LOUIS CABOT. Immature State of the Odonata, Part IV. E. L. MARK. Studies on Lepidosteus, continued. " On Araclinactis. A. AGASSIZ and C. O. WHITMAN. Pelagic Fishes. Part II. H. L. CLARK. The "Albatross" Hawaiian Echini. with 14 Plates. Reports on the Results of Dredging Operations in 1S77, 1878, 1879, and 1880, in charge of Alexander Aqassiz, by the U. S. Coast Survey Steamer "Blake," as follows: — A. MILNE EDWARDS and E. L. BOUVIER. The Crustacea of the "Blake." A. E. VERRILL. The Alcyonaria of the "Blake." Reports on the Results of the Expedition of 1891 of the TJ. S. Fish Commission Steamer "Albatross," Lieutenant Commander Z. L. Tanner, U. S. N., Commanding, in charge of Alexander Agassiz, as follows: — K, BRANDT. The Sagittae. The Thalassicolae. O CARLGREN. The Actinarians. R. V. CHAMBERLIN. The Annelids. W. R. COE. The Nemerteans. REINHARD DOHRN. The Eyes of Deep-Sea Crustacea. H. J. HANSEN. The Cirripeds. " The Schizopods. HAROLD HEATH. Solenogaster. W. A. HERDMAN. The Ascidians. S. J. HICKSON. The Antipathids. E. L. MARK. Branchiocerianthus. JOHN MURRAY. The Bottom Speci- mens. P. SCHIEMENZ. The Pteropods and Heteropods. THEO. STUDER. The Alcyonarians. The Salpidae and Doliolidae. H. B. WARD. The Sipunculids. Reports on the Scientiflc Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Com- manding, as follows: — R. V. CHAMBERLIN. The AnneUds. H. L. CLARK. The Holothurians. The Volcanic Rocks. The Coralliferous Limestones. S. HENSHAW. The Insects. R. VON LENDENFELD. The Silice- ous Sponges. Thff Ophiurans. •G. W. MtJLLER. The Ostracods. MARY J. RATHBUN, The Crustacea Decapoda. G. O. SARS. The Copepods. L. STEJNEGER. The Reptiles. C. H. TOWNSEND. The Mammals, Birds, and Fishes, T. W. VAUGHAN. The Corals. Recent and Fossil. PUBLICATIONS OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. There have been published of the Bulletin Vols. I. to LIV.; of the Memoirs, Vols. I. to XXIV., and also Vols. XXVI. to XXIX., XXXI. to XXXIV., XXXVI. to XXXVIIL, XLL, and XLIV. Vols. LV. to LVIII. of the Bulletin, and Vols. XXV., XXX., XXXV., XXXIX., XL., XLIL, XLIIL, XLV. to XLVIII. of the Memoirs, are now in course of publication. The Bulletin and Memoirs are devoted to the publication of original work by the Officers of the Museum, of investigations carried on by students and others in the different Laboratories of Natural History, and of work by specialists based upon the Museum Collections and Explorations. The following publications are in preparation: — Reports on the Results of Dredging Operations from 1877 to 1880, in charge of Alexander Agassiz, by the U. S. Coast Survey Steamer "Blake," Lieut. Commander C. D. Sigsbee, U. S. N., and Commander J. R. Bartlett, U. S. N., Commanding. Reports on the Results of the Expedition of 1891 of the U. S. Fish Commission Steamer "Albatross," Lieut. Commander Z. L. Tanner, U. S. N., Com- manding, in charge of Alexander Agassiz. Reports on the Scientific Results of the Expedition to the Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. Moser, U. S. N., Commanding. Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific, in charge of Alexander Agassiz, on the U. S. Fish Commission Steamer "Albatross," from October, 1904, to April, 1905, Lieut. Com- mander L. M. Garrett, U. S. N., Commanding. Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. Contributions from the Geological Laboratory, Professor R. A. Daly, in charge. These publications are issued in numbers at irregular intervals. Each number of the Bulletin and of the Memoirs is sold separately. A price list of the publications of the Museum will be sent on appli- cation to the Director of the Museum of Comparative Zoology^ Cambridge, Mass. ^\%\ Bulletin of the Museum of Compa,rj.tive Zoology AT HARVARD COLLKGl:?. Vol. LVIIl. No. 7. MAMMALS FROM THE BLUE NILE VALLEY By Glover M. Allen. CAMBRIDGE, MASS., U. S. A.: PRINTED FOR THE MUSEUM. July, 1914. Reports on thE SclEiv''iitic Results of the Expedition to the East- ern I^ROViciLfPACiMC, IN CHARGE OF ALEXANDER AgASSIZ, BY THE U. S. Fish Commission Steamer "Albatross," from October, 1904, TO MAKCii, 1605,1 lilECJTENANT CoMMANDER L. M. GaRRETT, U. S. N., Commanding, published or in preparation: — A. AGASSIZ. V.6 General Report on the Expedition. A. AGASSIZ. I.i Three Letters to Geo. M. Bowers, U. S..Fish Com. A. AGASSIZ and H. L. CLARK. The Echini. H. B. BIGELOW. XVI. le The Medusae. H. B. BIGELOW. XXIII.s' The Sipho- nophores. H. B. BIGELOW. XXVI.^e The Cteno- phores. R. P. BIGELOW. The Stomatopods. O. CARLGREN. The Actinaria. R. V. CHAMBERLIN. The Annelids. S. F. CLARKE. VIII.s The Hydroids. W. R. COE. The Nemerteans. L. J. COLE. XIX.19 The Pycnogonida. W. H. DALL. XIV." The Mollusks. C. R. EASTMAN. VII.' The Sharks' Teeth. S. GARMAN. XII. 12 The Reptiles. H. J. HANSEN. The Cirripeds. H. J. HANSEN. XXVII.27 The Schi- zopods. S. HENSHAW. The Insects. W. E. HOYLE. The Cephalopods. W. C. KENDALL and L. RADCLIFFE. XXV.". ' The Fishes. C. A. KOFOID. III." IX.9 XX. =0 The Protozoa. C. A. KOFOID and J. R. MICHENER. XXII." The Protozoa. C. A. KOFOID and E. J. RIGDEN. XXIV.21 The Protozoa. P. KRUMBACH. The Sagittae R. VON LENDENFELD. XXI.^i The Sihceous Sponges. The Holothurians. The Starfishes. The Ophiurans. G. W. MiJLLER. The Ostracods. JOHN MURRAY and G. V. LEE. XVII." The Bottom Specimens. MARY J. RATHBUN. X." The Crus- tacea Decapoda. HARRIET RICHARDSON. II.2 The Isopods. W. E. RITTER. IV.i The Tunicates. ALICE ROBERTSON. The Bryozoa. B. L. ROBINSON. The Plants. G. O. SARS. The Copepods. F. E. SCHULZE. XI.n The Xenophyo- phoras. H. R. SIMROTH. The Pteropods and Heteropods. E. C. ST ARKS. XIII." Atelaxia. TH. STUDER. The Alcyonaria. JH. THIELE. XV." Bathysciadium. T. W. VAUGHAN. VI. « The Corals. R. WOLTERECK. XVIII.is The Am- phipods. 1 BlUl. M. C. z.. Vol. 2 Bull. M. C. z., Vol. 8 Bull. M. C. z.. Vol. « Bull. M. C. z.. Vol. 5 Mem. M. c. z., Vol. 6 Bull. M. c. z.. Vol. 7 Bull. M. c. z., Vol. s Mem. M. c. z., Vol. 9 Bull. M. c. z.. Vol. lo Mem. M. c. z. Vol. n Bull. M. c. z. Vol. 12 Bull. M. c. z. Vol. 13 Bull. M. c. z. Vol. " Bull. M. c. z. Vol. 15 Bull. M. c. z. Vol. " Mem M. c. z. Vol. 1' Mem. M. c. z. Vol. 18 Bull. M. c. z. Vol. 19 Bull. M. c. z. Vol. 20 Bull. M. c. z. Vol. 21 Mem M. c. z. Vol. 22 Bull. M. c z. . A''ol. 23 Mem M c z. , Vol. 2* Bull. M c z. , Vol. 25 Mem M c z. , Vol. 25 Bull. M c z. , Vol. =■ :Mem M c z. , Vol. XLVI.. No. 4, April, 1905, 22 pp. XLVI., No. 6, July, 1905, 4 pp., 1 pi. XLVI., No. 9, September, 1905, 5 pp., 1 pi. XLVI., No. 13, January, 1906, 22 pp., 3 pis. XXXIII., January, 1906, 90 pp., 96 pis. L., No. 3, August, 1906, 14 pp., 10 pis. L., No. 4, November, 1906, 26 pp., 4 pis. XXXV., No. 1, February, 1907, 20 pp., 15 pis. L., No. 6, February, 1907, 48 pp., 18 pis. XXXV, No. 2, August, 1907, 56 pp., 9 pis. LI., No. 6, November, 1907, 22 pp., 1 pi. LII., No. 1, June, 1908, 14 pp., 1 pi. LIL, No. 2, July, 1908, 8 pp., 5 pis. XLIII., No. 0, October, 1908, 285 pp., 22 pis. LIL, No. 5, October, 1908, 11 pp., 2 pis. XXXVII., February, 1909, 243 pp., 48 pis. XXXVIIL, No. 1, Jmie, 1909, 172 pp.. 5 pis., 3 maps. LII., No. 9, June, 1909, 26 pp., 8 pis. LIL, No. 11, August, 1909, 10 pp., 3 pis. LIL, No. 13, September, 1909, 48 pp., 4 pis. XLL, August, September, 1910, 323 pp., 56 pis. LIV., No. 7, August, 1911, 38 pp. . XXXVIIL, No. 2, December, 1911, 232 pp., 32 pis. , LIV., No. 10, February, 1912, 16 pp., 2 pis. . XXXV., No. 3, April', 1912, 98 pp., 8 pis. . LIV, No. 12, April, 1912, 38 pp., 2 pis. . XXXV., No. 4, July, 1912. 124 pp., 12 pis. Bulletin of the Museum of ComparatiTe ZoUlogj AT HARVARD COLLEGE. Vol. LVIII. No. 7. MAMMALS FROM THE BLUE NILE VALLEY. By Glover M. Allen. CAMBRIDGE, MASS., U. S, A.: PRINTED FOR THE MUSEUM. July, 1914. Xo. 7. — Mamvials from ihe Blue Nile Valley. By Glover M. Allen. Ix January and February, 1913, I accompanied Dr. J. C. Phillips on his expedition up the Blue Nile and the Dinder River in the interests o!' the ]\Iuseum. A considerable effort was made to collect the birds (see Bull. M. C. Z., December, 1913, 58, p. 1-2S) and mam- mals of the region, and Dr. Phillips has generously left to me the working out of the latter. Our route lay along the Blue Nile, from Sennar, where our real start was made, to Singa, the present seat of government for Sennar Province. At this point we crossed to the north bank, for the south bank is a game reserve, and proceeded along it to the Abyssinian border, stopping at Fazogli, an outlying ' gebel ' of the Al)yssinian foothills. We later retraced our steps to Abu Tiga, and thence crossed over to the Dinder, an affluent of the main river, that becomes partly dry in the rainless season. The upper portion of this river seemed to have been very little disturbed, and large game was abundant and very unsuspicious. Along the Blue Nile, however, and on the lower parts of the Dinder, the native population is increas- ing and there is much travelling up and down along the river banks. On the Blue Nile especially, parties of Arabs and negroes are con- stantly passing, and English officials make their rounds between Singa and Roseires or other points. With the increase of native population, the clearing of the land, and disturbance incident to human occupation, the large game must inevitably be gradually driven back or exterminated by hunting. It is generally believed that the native population of the Sudan, during the time of the Mahdi and his successor (1883-1898) was reduced through war, famine, and disease about 75%, amounting to the almost total extinction of the inhabitants along the Rahad and Dinder, as well as on the Blue Nile, so that many of the villages marked on the older maps no longer exist. This no doubt has been favorable for the increase of large game in later years. On these rivers now, however, the habitations are being^ reestablished gradually, and population will doubtless reclaim the country in time. It therefore has seemed worth while to record the more striking facts we noted concerning the habits and distribution of the larger mammals, for they must eventually be much reduced or destroyed altogether. A few species seem better adapted to survive 306 bulletin: museum of comparative zoology. than others and these, on the Blue Nile, for example, already show through their difference of habits, compared to their congeners of the upper Dinder, an adaptation to the changing conditions. The entire country up to the Abyssinian border is monotonously flat, and covered largely with an open forest of thorn trees among which the red-barked gum-arabic tree is conspicuous. A very few Fig. 1. — Sketch map of the Blue Nile Valley. small and isolated hills or 'gebels' project here and there abruptly from the plain, and alone break its monotony. The Blue Nile has cut a channel through this broad plain, but so steep are its banks for many miles in succession, that access to the water is difficult, and hardly to be obtained except where gullies, cut down during the torrential rains ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 307 of the wet season, afPorcl a passage. As settlements increase along the rivers, the native villages are planted at such spots, termed ' mesharats,' or "places where one can get down to the water." Since the large mammals are also dependent on these for reaching the water, the settlements result in driving them farther and farther away where there are 'mesharats' at a distance from habitations, with their accompaniment of droves of cattle, yelping dogs, and native hunters. The luxuriant growth of tall grass that springs up after the summer rains becomes exceedingly dry by late autumn, and the natives set fire to it and burn the country for many hundreds of square miles. The soil itself becomes transformed from a mass of sticky mud in the wet season to a hard baked or a powdery condition, often much cracked and very difficult for walking. Such unfavorable conditions appear to have had a direct influence in reducing the ground-living species to a minimum, so that it was very hard to obtain small mammals, and even in comparatively sheltered places the number of species was disap- pointingly few. According to local report, there is much more large game along the Blue Nile during the wet season and just previous to it, in April and May, when the drying up of the smaller and remoter pools forces the animals to seek the main stream. The rank growth of vegetation during the summer rains also causes a more general dis- persal. There has been but little collecting done in the area covered, though travellers have from time to time sent specimens to Europe. As long ago as 1842, Sundevall published descriptions of mammals obtained in Sennar by the Swedish traveler Hedenborg, but as then used, Sennar was a somewhat indefinite term applied to the country between the White and the Blue Niles. Ruppell and Heuglin later did much exploration in northeastern Africa, including journeys into the Sudan. They gave names to many of the species whose range includes the Blue Nile country. What has since been done in the study of the mamma- lian fauna of the region has been of fragmentary nature, and consists chiefly of reports on occasional specimens sent by Europeans to the museums of England and Germany. In 1898, Lord Lovat's expedi- tion crossed from southern Abyssinia to the Blue Nile Valley, and obtained a few specimens from the latter region, including a. new multimammate mouse, described by de Winton (1900). Captain S. S. Flower, of the Gizeh Zoological Gardens has several times l)een in the region to obtain living animals for the splendid collection under his charge. Mr. A. L. Butler, head of the Game Preservation Depart- ment of the Sudan, also knows the country well and has sent many specimens of birds and mammals to the British Museum. 308 bulletin: museum of comparative zoology. In general the mammalian fauna may be said to be typically African, with almost no trace of Eurasian species. It is a continuation of that of the upper Nile, though rather more reduced, and in the region cov- ered, quite without any of the desert species found in the Saharan sands to the north and northwest. The list of species observed follows. Syncerus aequinoctialis (Blyth). Nile Valley Buffalo. Bubalus caffer aequinoctialis Blyth, Proc. Zool. soc. London, 1866, p. 372. Bubalus azrakensis Matschie, Sitzb. Ges. naturf. freunde, Berlin, 1906, p. 169. In his review of the African buffaloes Matschie describes Bubalus azrakensis as a new species; it is based upon an imperfect skull from Roseires on the Blue Nile. He says that it belongs to those forms in which the horn is strongly bowed downward and differs from all the other species in that the inwardly bent tips of the horns turn suddenly back at the ends. This appearance is sho\\Ti in his photographic figure, in which, however, one of these tips is broken oflP. Moreover, as the figure shows, the skull is that of an immature animal in which the basal portions of the horns are unsolidified and have not been preserved, although the spread is 84 cm., a fairly large size for Nile Valley animals. The horns of three old bulls shot by Dr. Phillips on the Dinder River, are heavy and massive, the bases very broad, but not joining medially on the forehead, nor are they convex in this region as in the caffer type, but flattened, ridged, and broadly exca- vated. Their downward sweep reaches only about to the level of the orbit and the tips are blunt and rather short, due in part to wear. Cotton (1912) says that the horns of cows have a deeper curve than those of the bulls and are not so wide. The long points, backwardly turned, of Matschie's azrakensis seem more like an individual varia- tion in an immature animal. In view of these facts, it does not appear that the Buffalo of this region is satisfactorily distinguished from aequinoctialis of the White Nile, so that it is best at present to use this latter name to include the Buffalo of the Blue Nile as well. The generic name Syncerus was revived in 1911 by Hollister to distinguish the African Buffalo from the Water Buffalo — Bubalus. The following measurements of Dr. Phillips's specimens were made in the field : — 1 2 3 mm. mm. mm. 2615 2470 2400 785 700+ 795 595 610 615 290 290 1680 1570 1660 1150 1145 190 180 170 135 140 125 140 160 160 110 125 110 745 850 880 210 230 210 ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 309 Nose to root of tail Tail (from anterior base to tip) Caicaneum to tip of hoof Ear from meatus to tip Standing height at shoulder Half girth back of fore leg Fore hoof, length of under side " " greatest breadth Hind hoof, length of under side " " greatest breadth Greatest expanse of horns, outside Greatest width of basal expansion of horn From these measurements it appears that the animal with the smallest spread of horns was the largest in body. The one with the broadest spread, however, (880 mm. = 34.5 inches) did not have the broadest base. The greater size of the anterior hoofs is also apparent; and is greatest in the largest-bodied specimen with the least spread of horns. Buffalo are now rare on the Blue Nile, at least along the north bank where our route lay. The only place where we learned of their presence was near a small native village called Omdurman, a few days' journey below Roseires. Here apparently was a small herd of perhaps eight or ten, that came almost nightly to the edge of a large marsh or to the vegetable gardens of the natives. They were very wary and dur- ing the day were not to be discovered, for they frequented the thickest cover along the river. As the natives are without firearms, the Buffalo have little to fear from them, though with shouts and fire- brands at night the men often frighten them from the growing crops. Passing sportsmen, or English officials, however, sometimes stop to hunt here. Cotton (1912) notes Buffalo at El Garef. Matschie's specimen of B. azrakensis is said to have come from Roseires, but may not have been shot in that immediate vicinity. It was not until we reached the Dinder that we found Buffalo in any numbers. As the district was closed for a time, we were obliged to retrace our steps from Roseires several days' journey down stream before we were allowed to cross over, a two days' march to the Dinder. This stream goes partly dry in the rainless season, so is much less disturbed and only very sparingly settled by natives. At El Kuka 310 bulletin: museum of comparative zoology. we first found Buffalo tracks, but these indicated only a few scattered animals. Continuing several days* journey to the vicinity of Um Orug, a large island in the stream bed, we finally came upon Buffaloes in such numbers as are hardly to be found elsewhere in Africa at the present day. At Khor Galegu was the last native village, and at some distance above this began a series of so-called ' meres,' which are great marshy areas often a mile long, and even at this dry season (February) moist or even boggy, with a rank growth of high grass, now largely eaten down by the wild game. For to these places re- sorted the large ruminants for miles around. It was near such a meadow, near Um Orug that we encountered a herd of some 250 Buffalo as they came at sunset to drink at a large pool in the river bed. Later we saw what was no doubt the same herd on a great 'mere' below this spot. On a 'mere' near a camping spot called Beit el Wahsh, we saw a second herd of about 100 old and young, and near a camp El Abiad, a herd of some sixty or more on a similar ' mere.' A very large old bull was seen here, that seemed to have been driven from the herd and was at the opposite side of the 'mere.* This and two other old bulls that were found together on another 'mere' far from any herd, fell to Dr. Phillips's rifle. They were all much battle-scarred, and one had lost an eye, and its ears were badly torn. The appearance of a herd of Buffalo at a distance is highly charac- teristic. They mass closely together, and their great black bodies form a solid rank, whose outline is hardly broken by the heads and horns, as these are carried nearly on a level with the back. The small White Egret often feeds close among the herds. At Abiad we saw a large flock of these birds, their white plumage in strong contrast to the black bulk of the great beasts. Strepsiceros strepsiceros CHORA (Cretzschmar). Northern Greater Kudu. Antilope chora Cretzschmar, Riippell's Atlas reise nordlichen Afrika. Saugeth. , 1826, p. 22. Pocock (1905) has proposed to distinguish the Greater Kudu of northern Africa as a distinct race from that of South Africa, and revives Cretzschmar's name for it. It is readily distinguished by its fewer white body stripes. Unquestionably the Kudu is the finest of the antelopes of the Nile ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 311 Valley. Its wariness, its love for hilly or broken ground, its keenness of sense, and its handsome appearance make it by far the most note- worthy of the large game mammals of the country it inhabits. Its present distribution along the Blue Nile is very interesting, as it frequents the narrow and intermittent strip of broken groimd a short distance back from the river where torrential streams have worn little valleys or ' khors ' in a soil locally harder or more gravelly than most of the level plain of the great river. Here there is more or less good cover, clumps of thorn bush, tall grass, or vines, which added to the irregular nature of the ground, forms a tolerable shelter. The method of hunting is to follow the track and by keen watching and silent following, to discover the animal before he is aware of the pursuit. Owing to the somewhat dense cover, however, or the dry grass and twigs, this is a difficult matter. It is usually the case, that the bulls are apt to be solitary and are much more difficult to approach than the cows, which often go in bands of three or four. We startled a company of three near Gebel Maba, and were told of a band of four being seen near Roseires. The former is a favorite haunt, an isolated and irregular hill, very stony, though with few large boulders, and covered with thorn trees. Mr. Savage at Roseires had lately taken a bull with fine head a few miles back from that post, and said that it was accompanied by a cow. Dr. Phillips at one time found a bull and calf together near Magangani, and spent much time following others at various points as far down the river as the neighborhood of El Mesharat, where he heard one giving its characteristic bellow. This sound is made by both l)ulls and cows. Near Magangani, Dr. Phillips was once watching a Kudu cow as she was lying dowTi, a hundred yards distant. Presently she rose to her feet and began to bellow at regular intervals of five seconds. As described by Dr. Phillips the sound is a single low explosive puff, like that of a distant freight engine heavy laden. This bellow he several times heard while follow- ing a Kudu track but the wary antelope always kept ahead just out of vision. The bleaching skulls and skeletons of male Kudu are not infrequently found, but those of cows much less often. Some perhaps are killed by lions, or wounded by hunters and lost. Certainly how- ever, there are comparatively few bulls left along the Blue Nile. On the Dinder River, the Kudu is practically absent except in a small stretch just below Um Orug, where Dr. Phillips heard the characteristic bellow and saw tracks. No doubt there are Kudu above this point but we did not go farther. There is much variation in the angle at which the horns come off 312 bulletin: museum of comparative zoology. from the skull, as well as in the openness of coilmg of the separate horns. Some are more open and divaricating, others a sHghtly closer spiral and the horns nearer together. The native hunters consider that there are two sorts of Kudu, which really are but the extremes of these two types. They call the former ghazdwi, the latter kardri, and be- lieve each animal is the property of some spirit who marks his animal that he may be known to his owner. The slit ear of one specimen is thus taken to be such a mark. The hoofs are remarkably delicate in proportion to the size of the animal and to an experienced eye, make a characteristic track. In the stomach of an 11-foot crocodile killed above Bados, Dr. Phillips found what seemed to be the hoofs of a Kudu. Tragelaphus decula (Riippell). Abyssinian Bushbuck. Antilope decula Riippell, Neue wirbelth. fauna Abyssinien. Saugeth., 1835, p. 11, pi. 4. A few Bushbuck may be found along the Blue Nile at Abu Zor and beyond, but the^^ are uncommon and solitary in habits. At El Garef we heard in the early evening what our native hunters assured us was their, curious sharp bark of alarm, reminding one of a small terrier. Cotton (1912) mentions finding Bushbuck at Bados and Magangani. It is apparently less common on the Dinder, for we met with it but once, at Um Orug where Dr. Phillips obtained a young male. Egocerus equinus bakeri (Heuglin). Baker's Roan Antelope. Hippotragus bakeri Heuglin, Nova acta Acad. Leop. Carol., 1863, 30, art. 2, p. 16, pi. 2, fig. 6. This fine Antelope is still to be found in small numbers on the north bank of the Blue Nile in the vicinity of Bados. They are shy, how- ever, and seem to use much caution in approaching the river to drink. This they do at some very early hour, and are far back in the dry thorn bush by daybreak. At Magangani, near Roseires, a small herd passed within a stone's throw of our encampment during the night, on the way to the water. Cotton speaks of finding one at Gebel Maba, some distance above Roseires, but there seem to be few beyond that ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 313 point. At Bados we spent a day hunting the Roan, with a skih'ul native tracker. The animals were well back from the river, and after about an hour's walk we reached their country, and spent some hours following tracks on the powdery "cotton soil" in the thorn scrub. The tracks were mostly of single animals or pairs, and we found where they had roamed about stopping here and there to bite off a green twig of a particular species of thorn, white-barked and with small obovate leaves. The Antelope were extremely shy and several broke away before we had even sighted them. Finally Dr. Phillips success- fidly stalked to within ninety yards of one lying apparently asleep imder a 'laloab' tree, at noon. But the watchful animal was quick to detect the motion of the binoculars, even at that distance and down wind, and leaped to its feet, a fine imposing creature. When startled at close range, the Roan as it Ijounds away makes a sound like a " sneezing cough." On the Binder, there are many more than on the Blue Nile. For some distance above the villages where the river bank is more or less travelled by Arab gum pickers and hunters, the Roan are shy, and their tracks, which we began to find at the camping spot. El Kuka, usually led straight back into the thorn scrub, so that it was fully a mile from the stream before the trails began to break up. Beyond the junction of the Galegu we saw many Roan. They had e\idently been little disturbed here and travelled in bands of as many as fifteen to twenty-five, taking no apparent precaution to avoid the river borders. Unlike the other antelopes, they seemed to avoid the open ' meres ' but were usually in the scattered tree growth, or the edge of the tall grass and bushes near the stream. They seemed to browse rather than graze. At Abiad several came to water at a pool of the Binder, in mid-afternoon, and it was interesting to see some drop to their knees to drink, though others drank standing. Owing to its wariness and its habit of retiring far back from the travelled river banks, this large species will no doubt continue to survive along the Blue Xile for some time longer. Cotton (1912, p. 53) believes that they drink only about twice a week, so are able to go a long way from water. He says they are still common on the Setit and the Atbara Rivers, in the uninhabited portions, but no longer exist on the Rahad. The stomach of one contained in the first compartment over a peck of the small twigs and leaves of a gray-barked thornl)ush, as well as a number of 'laboab' fruits, whose large stones are evidently masticated with the cud, instead of being regurgitated as with the smaller gazelle. 314 bulletin: museum of comparative zoology. Gazella soemmerringi (Cretzschmar). Ariel or Sommerring's Gazelle. Antilope soemmerringii Cretzschmar, RuppeU's Atlas reise nordlichen Afrika. Saugeth., 1826, p. 49, pi. 19. Of this species, Cotton (1912, p. 57) writes: "On the Atbara . . . . it was a rare animal ; but throughout the Setit it was very abundant, and on the Rahad, from a march or two above Hawata to the Abys- sinian border, the ariel were to be numbered only by thousands, and their presence obviously accounted for the number of lions. There were large herds on the Galegu and Binder, but not many of them, and on the Blue Nile I did not see a single specimen." It is the only species of gazelle that we found in all the country traversed. Cotton did not learn of its presence on the Blue Nile, but we saw a few back from the river near Bados, which appears to be the last remaining stretch of good game country on the north side of the river. This is no doubt because there is an area of marsh along the river which allows the animals to come to water without passing too close to villages. They must drink very early in the morning, for they are well back in the thorn scrub by daylight. On the south bank of the Blue Nile there are good numbers still, as we were informed by some officers of the Scots Guards, who obtained several heads there during our stay in the country. That side of the river is much less populated and is a reserve for use of officials only. In crossing from the Blue Nile to the Binder, from Abu Tiga, we saw a single bunch of three Ariel, but they are clearly very scarce in the region. It was not until we had proceeded some distance up the Binder that the Ariel began to appear. Near Ereif el Bik, a camp site by the bank, we saw a few coming from the water in early forenoon, and from this point on to Um Orug they were common, far outnumbering all the other antelopes. At times they were in sight nearly all day in smaller or larger bands ; f reqViently we started them in the forenoon at eight or nine o'clock coming from the water, and I have seen them come to drink as late as 12.30 P. M., for here they seemed to have been undisturbed for some while, and had lost much of their wariness of human kind. They are a most social species and gather into bands that number often fifty, seventy-five or a hundred approximately, of both sexes, and in early February the females were often accom- panied by young fawns. It was common to find single animals as ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 315 well, and these were usually old bucks. One which Dr. Phillips shot, showed many battle scars about the neck from the horns of some others of its kind, by which it had probably been driven from the herd. Once I saw a young buck butting playfully at the rear of one in front, and on another occasion Dr. Phillips had a good opportunity to observe their manner of fighting. Two bucks were seen fencing. They would lower their heads and catch each other's horns by the hook-like tips. Then followed a sort of tug-of-war in which one tried to pull the other about while their horns were thus interlocked. Sometimes they would l)utt at each other, and inflict sharp digs on the neck with the incurved tips of the horns. The chief food seemed to be grass, which was very closely grazed do\\Ti on the 'meres.' Away from these places there was very little green vegetation except bushy growth, l)ut everywhere the sprouting grass stalks were cropped off, and it was clear that green pasturage was none too plentiful for the big herds. The Ariel eat quantities of the date-shaped fruit of a species of thorn tree called the 'laloab,' which they pick up from the ground. This has a thin but juicy and rather acrid pulp with a large stone, enclosing a seed which is ground and eaten by the negroes. The stones appear to be regurgitated after the pulp has been digested, and it was common to find little heaps of half a dozen or so of these, quite cleaned, deposited on the ground. Our native hunters said that these were left by the Gazelles, after having been regurgitated, and though we did not actually see the process, there is no reason for doubting that this is the truth. These Gazelles seemed to be the most wary of the smaller antelopes. When feeding on the open with other grass-eating species, they were usually the first to take fright at our approach, and would move off, slowly at first, gathering sometimes into dense bunches like sheep, which they further simulated in their very whitish appearance. They are very conspicuous against the dark "cotton soil " or the burned areas, but among the dry and withered grass or on sand the contrast was less. They are constantly switching their tails from side to side, both when running or when standing, as though from sheer nervous- ness. I have seen the same habit in the Thompson's and Grant's gazelles. When surprised near the drinking places, they always seemed much concerned to get back from the belt of tall grass or shrubbery near the bank of the stream, but on reaching the more open thorn scrub, would stop to look about. Evidently they feared lions or leopards lying in wait at such places. Lions certainly kill good numbers, and we found the remains of several recent " kills." It was 316 bulletin: museum of comparative zoology. noticeable that most of these were youngish animals of small horns, no doubt the less experienced or less wary members of the herds. Occasionally aged animals are also killed, possibly because they are less able to escape through battle wounds or sickness. The result is therefore that in nature, the greatest mortality is among the youthful and inexperienced or among the aged and outworn. The finest specimens tend thus to be left to perpetuate the herd. It is worth noting that the effect of human game-protective laws is more or less the reverse, for the sportsman is usually content to let the poorer heads go, and to cull out those with the finest horns. In addition to lions, the Ariel evidently have much to fear from the crocodiles that lurk in all the large pools. In the stomach of one shot at Gosar, Dr. Phillips found horns of three Ariel, a doe and two small bucks, appar- ently. If possible the Gazelles will drink at a shallow pool in pre- ference to a large deep one, in which there are likely to be crocodiles. It would be interesting to know how active these Gazelles are by night. \Yhile marching by moonlight along the Dinder, we once came upon two that seemed to be grazing, and again in the dim light preceding dawn I found a few single animals moving about near the stream. The type locality of this species is the border of the Red Sea, but it has not yet been shown that the Ariel of the eastern Sudan is different, although two other races are described from more southern areas. Cervicapra bohor (Riippell). Bohor Reedbuck. Redunca bohor Riippell, Mus. Senckenbergianum, 1845, 3, p. 182. The Reedbuck is no longer common on the Blue Nile, and we met with it at but two places. El Mesharat and Bados. It is a most unsuspicious animal and no doubt one that will soon be much reduced in numbers. It has a way of standing broadside to the* intruder, the hind feet one in advance of the other, and with graceful head turned,, it sniffs the air and watches until certain that there is cause for alarm, when it bounds away with tremendous leaps. On the Dinder it was very common above El Kuka, and on the great open ' meres ' and along the grassy jungles by the stream bed they were found feeding through- out the day. They seemed to have been undisturbed here for a long period, and in contrast to their behavior on the Blue Nile, where they ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 317 had learned to keep under cover during most of the (hiyhght hours, they were extraordinarily tame. Unless the wind brought the taint of human scent, .they were almost without fear, but stood gazing within a few yards. On the Binder they were commonly in small groups, often an old buck with three or four does and once a younger buck On becoming alarmed the does would retreat at once leaving the old buck standing at gaze. We once came suddenly upon a young- ish animal that evidently had not seen us until it suddenly looked up from feeding a few yards to one side of the trail. At once it dropped flat upon the ground with head stretched out. We watched it a few moments, and as soon as we passed on it lifted its head to gaze after us, but remained crouching among the few stalks of tall grass that afforded not the slightest cover. Near Um Orvig I watched a Reedbuck as it came to w^ater, shortly after noon, with several Ariel. It drank much longer than they, stopping now and then to look about, but apparently quite uncon- cerned for the crocodiles, several of which lay a short distance off in the water. On one of the large open 'meres' we found Reedbuck active and apparently grazing by moonlight late in the evening. They were always the last of the antelope to take fright and run off when the caravan came out upon the 'mere' where they were feeding. We saw two large bucks, each with the tip of a horn broken off. It is possible that our specimens may be referable to the race cotioni but material is not at hand to settle this point. Johnson (1903) records killing a very large one on the Binder at Burraba in 1901, but we saw none so far down that river. CoBUS DEFASSA (Rlippell). Abyssinian Waterbuck. Antilope defassa Rlippell, Neue wirbelth. fauna Abyssinien. Saugeth., 1S35, p. 9, pi. 3. On the Blue Nile the Waterbuck is greatly reduced in numbers and no doubt will be practically gone in the course of a few years. We saw almost nothing of it on this river, though Br. Phillips found a few near Adreiba above Roseires, and we were shown a fair head killed near the latter place by Mr. Savage, then acting chief of the district. Cotton, however, in 1911, found Waterbuck at Bados, but if any are to be found below this region, they must be rare indeed. 318 bulletin: museum of comparative zoology. Quite different is it on the upper part of the Dinder. On reaching the stretches where the great open 'meres' begin, shortly below Um Orug, we found them really common. Cotton (1912) found them com- mon on the Galegu, but saw only a few on the Dinder, below the junc- tion of these two streams. He adds that they are not found on the Rahad, but are common on the Setit. We first found them a short distance above the Galegu, at Beit el Wahsh, where a few were feeding on a large ' mere,' and beyond this point we saw small numbers, usu- ally feeding in similar places. At Um Orug they were plentiful and remarkably unsuspicious. Dr. Phillips at one spot came upon a Waterbuck that allowed him to walk entirely around it at a short distance. Another that he obtained must have recently escaped from a lion as its back was deeply scratched and its belly so injured that the intestines protruded through the open wound. Shortly above the Galegu junction we saw a lion stalking an old Waterbuck in the open sandy bed of the river at about midday. On one large 'mere' at Um Orug, Dr. Phillips saw ninety-seven Waterbuck at one time, quietly feeding, and later that evening we found others there, grazing by the half moonlight. This with the Reedbuck was usually the last of the antelopes to take fright when several species were feeding together. During the heat of the day they are apt to rest under the shelter of the thorn trees, and it is common to see them in small parties consisting of a buck and two or three does. Ourebia MONTANA (Cretzschmar). Abyssinian Oribi. Antilope montana Cretzschmar, Riippell's Atlas reise nordlichen Afrika. Saugeth., 1826, p. 11, pi. 3. In his original description of this species, Cretzschmar gives its known range as Bahr-el-Abiad and the mountains about Fazogli (spelled "Fazuglo"). The latter are merely hills, however, so that the name " 7nontana" is somewhat misleading. This is the common antelope along the Blue Nile and is called by the Arabs "ghazal." In many of its habits it corresponds to our Virginia Deer. It inhabits the edge of the tall grass jungle along the river bank, or the bushy tangles in which it finds a safe retreat. W'e also met with it on the slopes of the 'gebels' or hills. It is watchful and resourceful, yet hardly to be considered shy, so that it seems well adapted to survive ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 319 in the presence of civilization, and will doubtless continue in the hind long after the other species of antelopes have been exterminated. The country between Sennar and Singa is so travelled and cultivated that we saw none on that part of the road, but bej^ond the latter town we saw them almost daily. In the early morning they are about before sunrise feeding, but usually are less in e\idence after six or seven o'clock, especially in the neighborhood of villages, for they retreat to cover and come out again towards evening. Yet we often saw them even at midday, standing in the tall grass, gazing attentively at us as we passed. Often they would stand thus watching till we were out of sight, but if alarmed by a suspicious movement or a too close approach they scurried off at once into the thick cover. Along the Blue Nile we saw them frequently in pairs, and singles, and a good number were accompanied by little fawns in January. Their curiosity is consider- able and almost always causes them to stop, after the first dash, and stand broadside on watching intently the object of their suspicion, and thus affording the hunter an easy shot. Away from the river there were but few Oribi, and in crossing to the Binder we saw but a single one not far from a small and partly dried waterhole. Along the Binder, Oribi were abundant and we often saw small troops of four or five. Here they were little disturbed and surprisingly tame, allowing us often to approach within a very short distance. They frequented the edges of the open 'meres' with the other antelope, throughout the day. Their cry of alarm is a sharp whistled " phee-u," not so hoarse, it seemed to me, as the somewhat similar whistle of the Reedbuck. It is often difficult to distinguish Oribi from small or hornless Reed- bucks, especially as the two occur together along the edges of the grass jungles, but there are several very characteristic traits that serve to identify the two. In running away the Reedbuck holds its tail tightly down between its legs, whereas the Oribi holds its tail stiffly erect, exposing the blackish skin about the anus. Its gait is also stiffer, with a sort of bobbing up and down of the hind quarters as it scurries along, whereas the Reedbuck has a much freer gait, and often takes beautiful deer-like bounds, fore feet out in front, hind feet straight out behind, as it clears some obstructing bush. In reporting on the mammals obtained by Lord Lovat's expedition from southern Abyssinia to the Blue Nile, de Winton (1900, p. 8-1) states that specimens of the Beira Antelope (Borcotragus) were brought back. According to Lord Lovat " the Beira Antelope is com- mon all down the Blue Nile to Roseires; it inhabits the slopes leading 320 bulletin: museum of comparative zoology. down to the river-bed, and is also seen on the barer hill-tops." The presence of this hill-loving species in the Blue Nile Valley west of the Abyssinian hills would certainly be extraordinary, and I cannot but think that on reaching this low flat country Lord Lovat mistook the Oribi or the Duiker for the Beira which he had found in the higher land through which he had just passed. At all events we found no sign of it between Roseires and Fazogli during our trip. Cephalophus abyssinicus Thomas. Abyssinian Duiker. Cephalophus abyssinicus Thomas, Proc. Zoql. soc. London, 1892, p. 427. Specimens from the eastern Sudan are currently referred to this species. We met with it in the Blue Nile Valley only, and in but few places. Cotton (1912) records seeing one at Bados and it is likely that this is about as far north as it occurs on the Blue Nile. Above this point we saw a few at Magangani, and near Gebel Maba, and some numbers near Fazogli. This is a very sedentary animal, and we repeatedly found what were presumably the same individuals near the same thickets day after day. At Magangani we saw a few along the edge of a great sea of elephant grass between the river and some undulating ridges, but at Fazogli they frequented a considerable area of alternating ridges and small gullies, which with their thickets and clumps of grass or small palms were admirable hiding places. Dr. Phillips spent much time here observing them, and found them most crafty and watchful. They were usually seen in the early part of the day singly or in pairs, and had a way of hiding in clumps of vegetation to watch the intruder or slinking adroitly off under cover of an inter- vening bush if followed. Those living near this native village were no doubt much hunted and had become extremely adept at keeping out of sight. Damaliscus tiang (Heuglin). Tiang Hartebeest. Damalis tiang Heuglin, Nova acta Acad. Leop. Carol., 1863, 30, art. 2, p. 22, pi. 1, fig. la, b. According to Cotton (1912, p. 55) this handsome antelope is not now to be found on the Setit, Atbara, or Rahad Rivers, although it is ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 321 plentiful on the Galegu. We found them rare on the Blue Nile, and saw them only in a few places, near Bados and ^lagangani, below Roseires. They are more or less hunted here by passing sportsmen and have become shy and watchful. They usually go in small herds of ten or less and come to water at a few places removed from the villages. After drinking they at once leave the river and are some miles back in the thorn bush by daylight. On the upper Dinder, where they seemed to have been unmolested for some time, their behavior was c^uite different. On our way up this river we first came upon them near a loop of the stream called Ereif el Dik (the cock's comb, in allusion to the sinuous course of the stream), where a small herd was started at noon from under some 'laloab' trees, whose date- shaped fruit they had been nibbling on the ground. But it was not imtil the region of the big open meadows or 'meres' was reached, at Beit el Wahsh and Abiad that they were found in numbers, while from this point to Um Orug they were very common. On one such 'mere' we estimated that nearly a thousand were in sight, feeding quietly in the open most of the day, while it was not uncommon to count seventy -five or one hundred on smaller 'meres.' Contrary to their habits along the Blue Nile, they seemed to be here under no restraint, and largely avoided the dry thorn bush, but fed on the grassy 'meres' most of the day. They were nevertheless watchful and were usually the first after the Ariel to take alarm, and to run off in a somewhat panicky way. Two female specimens collected here in mid-February contained each a large foetus. BUBALIS TORA RAHATENSIS Matschic. Eastern Sudan Hartebeest. Bubalis tora rahatensis Matschic, Sitzb. Ges. naturf. freunde, Berlin, 1906, p. 246. The type of this race came from Shunfar, a tributary of the Rahad, and its describer mentions a second specimen from about thirty miles southwest of Lake Tana, adding that it apparently is found on the entire middle Blue Nile, the Rahad and the Dinder. We were unable to discover any sign of the species on the Blue Nile, however, and if it now occurs along that stream, west of the Abyssinian boundary, it must be extremely rare. On the upper Dinder, there are a few, but they are scarce indeed in comparison with the Tiang. From Abiad to 322 bulletin: museum of compakative zoology. Um Orug we saw in all a fair number, usually in pairs, with other antelope on the great 'meres.' One herd of fifteen was deemed unusual. I came upon a fine lone bull drinking at a pool of the river an hour before noon. It seemed much astonished, but was not thoroughly alarmed until it got my scent, when with a loud explosive "oof" it bounded away. GiRAFFA camelopardalis (Linue). Nubian Giraffe. Cervus camelopardalis Linne, S,yst. nat., ed. 10, 1758, 1, p. 66. Thanks to governmental protection, Giraffe are still present in small numbers in parts of the Blue Nile Valley and on the upper Dinder. Mr. A. L. Butler of the Game Preservation Department said that they had very noticeably increased of late years. We saw none during our sojourn along the Blue Nile, but discovered old tracks in numbers some miles back from that stream; these were made during the rains when the ground was soft and were still (in January) deeply impressed in the sun-baked soil. The first locality where these tracks were seen was among the gum arable trees about Gebel Okalma, near El Mesha- rat. A few other tracks were found, some fairly recent, in crossing from the Blue Nile to the Dinder between Abu Tiga and Wad Shara Shara. On the upper Dinder we saw several small herds of Giraffe, usually on or near the open 'meres' or boggy areas overgrown with rank grass. A fine herd of ten was seen near Abiad, and later three others. Shortly below Um Orug we saw a herd of twenty-one and later another of twenty-five and after dark came upon a small herd that took headlong flight through the tall grass. Their chief enemy is the lion, and we several times came upon dead Giraftes that had evidently been killed by them. These were usually youngish animals with the epiphyses of the bones still separate. The lions do not eat the tough hide of the Giraffes but leave this carefully separated from the carcase, and even the vultures merely pick it clean. On a ' mere' near Abiad we found a Giraffe that seemed to have died from natural causes — an old and scabby-looking animal with no external wound apparently. The gathering vultures had only just commenced upon it. A few young Giraffes are caught alive yearly in this region by the natives, with government permission, to be sent to Cairo or elsewhere for zoological gardens. The natural gait of the Giraffe when walking, ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 323 is (like that of the camel) a pace — the two legs of one side acting together, but Nvhen frightened the herds go off in single file at a stiff gallop, their long necks held forward at an angle and undulating with a sinuous movement. Hippopota:mus .^iphibius Linne. Hippopotamus. Hippopotamus a7nphibius Linne, Syst. nat., ed. 10, 175S, 1, p. 74. This fine mammal is doomed to extinction in the Nile ere many years. Not only does he present an easy mark for hunters as he rests on a sandbar but on account of his occasional attacks upon small boats and the damage done to native crops, protection is not now accorded him and his destruction is even encouraged. F. L. James, writing in 1SS4, of "The wild tribes of the Sudan," says that at that time hippos were no longer plentiful north of Khartoum. At Berber there were still a few but they were hunted by the natives who watched for them nightly as they came from the river to feed on the growing crops. At the present time hippos are practically gone from the river above Khartoum, though Captain S. S. Flower told us that about 1908 the tracks of one were seen that had walked across the point at the junction of the White and the Blue Niles close to that city. This was most unusual even then, however, for in 1901 I. C. Johnson (1903) recorded that during a voyage up the Blue Nile from Khartoum, the first hippo seen was near the mouth of the Dinder some forty miles below Wad ^ledani. He supposed this to be about its northern limit at that time. We saw no hippos on the Blue Nile until well above Singa, at El Mesharat, where there were several basking on the mud flats in the middle of the stream. They have become very shy from constant persecution by sportsmen and others passing up and down to Roseires and no doubt will soon be nearly gone from this part of the stream. We several times found their well-worn paths up nearly precipitous banks into the grassy jungles along the river and frequently heard their loud guttural honking at night. At Bados one was caught by the natives in nooses set in its path. These were attached to large wooden floats, which discovered the animal's whereabouts to his cap- tors the following morning after it had retired to the stream. Four spearmen in a large wooden boat went leisurely forth to attack their captive, a rather small specimen, but full of fight. Previous to the 324 bulletin: museum of comparative zoology. attack we watched the animal for some time and found that it came very regularly to the surface for air at intervals of 3.5 minutes. The fight was short but furious, the men jabbing with their spears each time the enraged beast rose to attack the broadside of the boat. When at last it rose no more, the watchers on the bank shouted exult- ingly and one tM^anged a small harp in praise of the hunters. No hippos were seen at Roseires, the head of navigation for large boats, but we observed a few above that town near Adreiba. On the Dinder there are very few, at least on the upper portion. This is partly on account of the intermittent nature of the stream, though in the larger pools an occasional one is found. At Um Orug a few skulls of young animals were seen, from which the front teeth had been removed. W. B. Cotton (1912, p. 43) says there are still a few in theAtbara and Setit Rivers, but none at all in the Rahad. Phacochoerus africanus bufo Heller. Nile Warthog. Phacochoerus africanus bufo Heller, Smithsonian misc. coll., 1914, 61, no. 22, p. 2. Small numbers of Warthogs are still to be found along the Blue Nile and on the upper Dinder. Dr. Phillips shot one at El Mesharat and we met with a few others along the river to Roseires. On the upper Dinder we saw not a few, once a party of three large ones with four young. As noted by Cotton (1912) there seem to be few if any with large tusks in this region. Two skulls preserved agree with Heller's description of the Nile Valley Warthog, and, as he points out, differ from the East African race in the prolongation of the parietal portion and the nearly flat interorbi- tal region. DicEROs BicoRNis (Liuue). Black Rhinoceros. Rhinoceros bicornis Linne, Syst. nat., ed. 10, 1758, 1, p. 56. The Rhinoceros is nearly extinct in the eastern Sudan. In the days of Sir Samuel Baker they were plentiful on the upper Atbara and the Setit, but now apparently there are extremely few between the Nile and the Abyssinian border. It is worth recording therefore, that at ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 325 the present time they are quite gone from the Blue Nile, but a very few yet remain on the uppermost reaches of the Dinder River, about a day's march beyond Um Orug Island, as our native hunters told us. According to our Arab guide who had hunted this region, one was killed in 1911 on the 'mere' near El Abiad by a white hunter, who mistook it at night for a Buffalo. Beyond Um Orug, at a place called Hageirat, south towards the Abyssinian border a few are still to be found. The Rhinoceros is protected under the present game laws of the Sudan, but the few that survive are more or less in danger from poaching Abyssinians. Capt. Stanley S. Flower told us at Cairo that so far as he could learn there were probably not more than ten or a dozen rhinos left on the upper Dinder, and that these are probably not breeding for the natives report no tracks of young ones. Lydekker (Proc. Zool. soc. London, 1911, p. 958) recognizes the Black Rhino of Somaliland as distinct under the name somaliensis, but in the absence of specimens I cannot attempt to settle the identity of the Sudanese animals. Elephas africaxus oxyotis Matschie. Sudanese Elephant. Elephas africaniis oxyolis Matschie, Sitzb. Ges. naturf. freunde Berlin, 1900, p. 196. In reviewing the African elephants, Lydekker (1907, p. 398) con- siders that the form inhaljiting the Blue Nile Valley and western Abyssinia may stand as a valid race. It is characterized by Matschie as having a very long and pointed lobe at the base of the ear. The upper border of the ear is much rounded but the value of this char- acter is still under discussion. The tusks are rather small in this race, nardly above 60 lbs. Elephants were formerly common over the eastern Sudan, and have been much hunted for their ivory. Sir Samuel Baker's accounts of their pursuit and capture by the Arab hunters, mounted on agile ponies and armed only with a keen-edged sword, are familiar to readers of African travel. At the present time Elephants are practi- cally gone from the travelled region along the northeastern bank of the Blue Nile. I. C. Johnson, in 1901, hunted Elephant near the little village of Omdurman above Karkoj, and although a small herd of five was discovered, the animals were traveling and struck off toward the 326 bulletin: museum of comparative zoology. Dinder. The southeasterly bank of the Blue Nile is a semireserva- tion, where government officials only are allowed to hunt, and there is mvich less travel and native settlement. The same writer mentions that Elephants occasionally come to drink on this south bank at Zu- murka, nearly opposite from Karkoj, and opposite Abu Tiga and Om Bared, farther up. The only place where we learned of their presence was opposite Magangani, a few miles below Roseires. Here we heard them trumpeting and blowing water about one evening in -January, but were unable to see the animals. They still frequent the Dinder River. In 1901, I. C. Johnson found them at Durraba and shot one near there. On our journey up this river we first found their tracks and droppingS'Siii the dry river bed above that place at a camp site, Mesharat el Kuka. The spoor was old, however. From this point on up the river to Um Orug, our farthest camp, there was abundance of old sign, and many broken trees twisted oft' by the huge beasts. A poaching party of Abyssinians had killed an Elephant here two or three months before and the herd had evidently left the region; possibly they had crossed over to the Rahad, or as some of the native hunters supposed, they may have retired to a khor or dry water course to the south. The red-barked Acacia, whence the gum arable is obtained, is the favorite food tree of the Elephants in this region. We constantly came upon large trees of this species, often eight inches in diameter at two or three feet from the ground and twenty-five or thirty feet high, that had been broken down and the topmost twigs eaten. They are broken in a rather characteristic manner, at about two or three feet from the ground, and the trunk partly twisted off. Others are broken over and uprooted, and the topmost twigs chewed. Procavia butleri Wroughton. Butler's Hyrax. Procavia buileri Wroughton, Ann. mag. nat. hist., 1911, ser. 8, 8, p. 461. The type of this species was obtained by Mr. A. L. Butler at Gebel Fazogli, one of the foothills of the Abyssinian highlands on the south side of the Blue Nile. Mr. Wroughton, in describing it, records a second specimen from Gebel Ain on the White Nile. During our stay at Fazogli we obtained three specimens and saw a few others. They live in dens among huge boulders and though somewhat shy, have a curious way of appearing suddenly at the openings of their retreats,. ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 327 or frequently coming boldly out several feet from the entrance, where perched on a boulder they look about or give a characteristic sharp bark of two syllables at short intervals for some minutes at a time. Apparently they are much preyed upon by leopards and no doubt by other smaller Carnivora or predacious birds. Their habit of throwing aside all caution and bounding a few paces from their holes of a sudden is thus rather the more remarkable. At times, however, they show more concern for their safety, and if alarmed, will sit juotionless at the opening of the den for many minutes at a time. Again they may be seen to run a long distance from rock to rock, and then dive into a crevice. When convinced that no danger is near they delight to bask in the sun during the early forenoon, but commonly retire at about 9:30 or 10 o'clock in the morning. On one occasion, however, I saw three running rapidly among the loose boulders at 1 p. m. On the rocks where they are accustomed to bask and particularly at the entrance to their dens, are usually to be seen large accumulations of their droppings. In addition to those from Gebel Fazogli, I found a considerable colony on a large isolated rock peak, Gebel Okalma. This is in appearance an old volcanic neck, projecting steeply and abruptly from the plain, several days' march from the nearest of the Abyssinian foothills from which it is separated by many miles of low country that w'ould be utterly impassable for a Hyrax. The presence of these isolated colonies must therefore indicate that they have been long in the land, probably before the deposition of the loess that now covers the country. I could, nevertheless, detect no single character by which the Okalma specimens differed from those of Fazogli. No trace of these animals w^as to be found on a neighboring hill (Gebel Maba), which, however, was much less rocky, and afforded no suitable boulder heaps. Arvicanthis testicularis (Sundevall). Field Rat. Isomys testicularis Sundevall, Kongl. Svenska vet.-acad. Handl., for 1842, 1843, p. 221. This is the common Field Rat of the Blue Nile valley in the Sudan, and occurs generally throughout the country traversed from Sennar to Fazogli. Its favorite haunts are grassy fields, the borders of culti- vated grounds, or the open scrub of bushes, weeds, and small palms. It is practically a diurnal species, and was several times seen running 328 bulletin: museltm of comparative zoology. about in the hotter parts of the day. The specimens trapped were all taken in early morning or before evening. Hawks catch many of them. Sundevall's description was drawn from specimens collected on the White Nile by Hedenborg, and appears to apply well to the series from Sennar. The body measurements of adults are larger than he gives, however, for the average of three adults is : — head and body 163 mm., tail 149, foot 35, ear 19. Apparently A. abyssinicus does not occur west of the Abyssinian border. At all events, persistent trapping failed to discover it; nor did Lord Lo vat's expedition across Abyssinia find it farther west than Sellen and Goodur in the high country at the head of the Blue Nile. AcoMYS ciNERACEUS HeugUu and Fitzinger. Gray-footed Spiny IMouse. Acomys cineraceus Heugl. and Fitzinger, Sitzb. Kon. akad. wiss. Wien, math.- nat. cl., 1867, 54, pt. 1, p. 573. Two species of spiny mice were collected by the expedition. The one is a broad-footed, shorter-tailed animal, inhabiting all the low flat country of the Blue Nile Valley; the other is a slender-footed, longer-tailed species which we found only at Fazogli in the rocky hills which begin here at the Abyssinian border. The former I have referred to Heuglin's A. cineraceus; Heuglin's type locality is Doka, in eastern Sennar, between the Atbara and the Rahad Rivers. The original description is brief and refers to a figure pre\dously published by Heuglin. In his "Reise" (1877), however, he gives a more detailed accoimt, with measurements, which agree in all essentials with those of an immature specimen taken at Adreiba, a day's march above Roseires on the Blue Nile. We were fortimate in obtaining a second adult speci- men, much farther down the river at El Mesharat. Apparently it is a widely distributed species but was difficult to obtain m the dry and barren plains over which we journeyed. There can be no doubt that Heuglin's type was an immature animal, ha\'ing the entire dorsal area a smoky gray, palmg slightly at the sides. The feet he states are marked with the same color on their outer portion. The measure- ments given are:— head and body 3" 3'" (= 82.5 mm.), tail 2" 6'" (= 69 mm.), ear 6'" (= 12.6 mm.). Our immature specimen meas- ures:— head and body 78 mm., tail 67. The ratio of tail to head and ALLEN : MAMMALS FROM THE BLUE NILE VALLEY. 329 body is 83^ in Heuglin's specimen, 85% in our immature individual. In the adult the tail is relatively shorter — 73%; and the measure- ments of the fresh specimen were: — head and body 112, tail 82, hinfl foot 18, ear 15. The entire dorsal region from nose to base of tail is smoke gray, becoming pale day-color on the checks and sides of the body. The forearms and outer sides of the metacarpal and bases of the metatarsal areas are gray like the back, and the tail is similar above. The ventral surfaces, a spot below the eye, and at the base of the ear are white. Capt. Stanley S. Flower, of the Gizeh Zoological Gardens, generously presented an adult Acomys in alcohol taken June 19, 1912, at Eneikliba in Sennar Pro\ince, which is unquestionably the same animal. The short, broad hind feet and the relatively short tail (85 mm.) are equally characteristic. In the No\atates zoologicae, (1901, 8, p. 400) de Winton describes as new% Acomys ivithcrbi/i, type from Kawa, south of Khartoiun. He compares it with A. nubicus of Heuglin, from ^Middle Egypt, and men- tions specimens from Shendi and Gebel Auli in the Nile \'alley. It seems very close to cineraceus, with which it appears to agree in all essential characters, so far as the description goes. Possibly the two are identical, and ciucraccus should apply to the Acomys of the level country of this part of the Nile \'alley. Acomys hunteri de\Yinton. Hunter's Spiny Mouse. Acomys hunteri de Winton, Novitates zoologicae, 1901, 8, p. 401, footnote. Among the rock cre\nces of Gebel Fazogli, at the Abyssinian border, there occurred a second species of Acomys, which from descriptions alone, I am unable to differentiate from hunteri, the type of which came from the plains of Tokar, near Suakin, on the Red Sea. This is described as red fawTi above, white beneath, w^hich is practically as in an adult from Fazogli, except for the darker spines of the head and back. A younger individual is grayer dorsally, the sides pale ochra- ceous. The measurements given are: — head and body 105, tail 102, foot 17.5, ear 16. Our two specimens measure: — head and body 104, 101, tail 98, 96, foot 18, 19, ear 16, 16. It will be obser\ed that the tail is about 94 or 95 (in the tj^pe 97) per cent of the length of head and body, hence much longer than in cineraceus. It differs strikingly also in its slenderer feet, which are pure white instead of darker. 330 bulletin: museum of comparative zoology. The white spot at the base of the ear is not conspicuous. Compared with A. kemjn from British East Africa, these specimens are only a trifle paler, and externally hardly to be distinguished. One specimen was taken in a trap placed on a leaning stump some three feet from the ground. Mus (Leggada) tenella (Thomas). Blue Nile Harvest Mouse. Leggada tenella Thomas, Proc. Zool. soc. London, 1903, 1, p. 298. Three specimens were preserved from Magangani and El Garef, both within a few miles of Roseires, the type locality. Two of the specimens are immature and much darker o^'er the back than the other which is an adult. The type is said to have the fore legs en- tirely white, but in these two youngish specimens they are buffy like the sides of the body, and very pale buffy in the adult. The white spot at the outer base of the ear is very marked, whereas in the dark L. bella of British East Africa this spot is practically wanting. Several other specimens were trapped along the Blue Nile at El Garef, Magangani, Bados, among the thorn bushes and tall grass canes, but they were nowhere common. The adult female measured: — total length 116 mm., tail 54, hind foot 13, ear from meatus 10. Epimys macrolepis (Sundevall). Large-scaled Rat. Mus macrolepis Sundevall, Kongl. Svenska vet.-acad. Handl., for 1842, 1843, p. 218. The identity of Sundevall's Mus macrolepis is still a matter of some doubt, as indicated by Wroughton (1911, p. 460), and its author was himself uncertain whether or not it was the same as Riippell's Mus albipes. The type locality of the former is Roseires, and there can be no doubt whatever, from Sundevall's careful description, that his macrolepis is the common ground rat which we found all along our journey from Sennar to Fazogli on the Blue Nile, and wherever we trapped on the Dinder River. The name is based on the fact that the caudal scales seemed large, five to 5 mm., but in our dried speci- mens there are six to 5 mm. No doubt Sundevall made the measure- ment from alcoholics. Until it can be shown, therefore, that Mus ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 331 albipes is identical with the Blue Nile rat, Sundevall's name may stand for it. At El Garef we found a large colony of this rat among an open scrub growth of small dhoum palms, weeds, and bushes. They make well-worn runways from one clump of palms to another, or among the weed tangles, and live in holes dug in the ground in these shelters. They are apparently for the most part nocturnal. The measurements of an adult male of E. macrolepis from Gabardi, beyond Singa, are: — head and body 142, tail 149, hind foot 24, ear 19. This is a very brightly colored specimen, with a buffy suffusion over the entire upper surface, and with a buffy line in the middle of the belly. The pure buffy tips of the hairs of the sides make a distinct stripe in this species, from the nose to the ankle, bounding the white of the belly. Of Epimys azrek, a species of the multimammate group, the type of which also came from Roseires, we could find nothing. It may be at once distinguished by its smaller dimensions and by its pure white belly hairs, which latter in macrolepis are dark gray at their bases. Tatera robusta (Cretzschmar). Nile Valley Tufted-tailed Gerbille. Meriones robustus Cretzschmar, Riippsll's Atlas reise nordlichen Afrika. Saugeth., 1826, p. 75, pi. 29, fig. b. Wroughton (1906, p. 494) in his review of the members of this genus shows that Sundevall's Meriones murinus is probably the same as the M. robustus of Cretzschmar, the type of which is still in existence and is labeled " Ambukol, Nubien." The latter name Wroughton applies to the tufted-tailed gerbilles of the Nile Valley, and includes among his list of specimens in the British Museum, a single example from the Blue Nile, at Roseires, collected by Lord Lovat's expedition. We found this the common species all along the Blue Nile. It lives in tangled growth of grass, bushes, and small palms, the shelter of which it commonly shares with the native rat {Epimys macrolepis). Tatera flavipes, sp. nov. Buff-footed Gerbille. Type. — Skin and skull 14,491 M. C. Z., adult female, from Aradeiba, above Roseires, Blue Nile, Sudan. January 22, 1913. 332 bulletin: museum of comparative zoology. General Characters. — Size large; tail as long as head and body, not tufted. Dark hairs prevailing dorsally; backs of hands and feet buff. Upper incisors grooved; pterygoids slightly expanded proximally, their bases extending forv\^ard anterior to the posterior median edge of the palate. Description. — Top of head from nose to crown, nape, back, and dorsal surface of the tail a mixture of black and oehraceous buff, the former predominating. The individual hairs are slaty for the basal two thirds, then either black-tipped or with a subapical ring of oehra- ceous buff and a black tip; on the sides of the muzzle, cheeks, sides of body, forearms, and hind legs the black-tipped hairs become largely suppressed giving a nearly clear oehraceous buft' (Ridgway, 1886) tone to these parts. Metacarpal and metatarsal areas clear buff, toes- white. The ears are clothed with minute blackish hairs externally, and sparsely covered internally with short pale buff hairs. The entire ventral surface of the head and body (including the upper lips) and the limbs are covered with hairs white to their bases. The tail is sharply marked off by its oehraceous buff color on its basal half be- low; the rest of the under surface is darkened with short black hairs which predominate towards the tip. The terminal hairs are slightly the longest but do not form a tuft. Skull. — The skull is that of a mature animal but the teeth are only slightly worn. In this condition the middle lamina of the first upper molar shows a slight central contraction mark- ing off an inner and an outer portion. The upper incisors show a well-marked groove nearer the outer side, thus differ- ing from the liodon group which this species equals in size. Anterior palatal foramina 3 mm. long, reaching from the level of the center of the first molar to the back of the second. Posterior pala-- tal foramina reduced to two minute per- forations just posterior to the level of the last molars. The conformation of the pterygoids is different from that of other species to which I have had access. Their bases are slightly divergent, and included within a notch in the palatal bones between the median projecting point and the posterolateral extensions, which are rather Fig. 2. — Tatera flavipes, palatal region. X 3. ALLEN: MAMMALS J"IlOM THE BLUE NILE VALLEY. 333 broader than in T. rohusta. Distally the club-shaped end of the pterygoid is strongly in contact with the large audital bulla. Measurements. — The type measured in the flesh: — head and body 171 mm., tail 172, hind foot 40.5, ear from meatus 23. Skull: — occi- pitonasal length 44, basal length 37.6, palatal length 24.4, nasals 17, zygomatic breadth 22.5, incisive foramina S, audital bulla 12.3 X 7.2; upper molar series (alveoli) 8, lower molar series (alveoli) 8, upper diastema 12.5. Remarks. — This large species seemed rare as we obtained but the single specimen. It was trapped in grass and bushes on the edge of a native grain field. In ^Yroughton's key to this genus (190b), it would come under the second Section, " A. Tail not appreciably longer than head and body." It seems to show no very close relation to either liodon or vallda, the two largest species of this section. Its large size, untufted tail equalling head and body, grooved upper incisors, dark dorsal area, buffy feet, and peculiar shape of the pterygoids are char- acteristic. At Fazogli, on the south side of the Blue Nile we obtained a second species of Tatera with untufted tail, which likewise seems undescribed. It may be known as Tatera soror, sp. nov. Lesser Blue Nile Gerbille. Tijpe.— Skin and skull 14,492 M. C. Z., adult female, from Fazogli, Blue Nile, Sudan. January 16, 1913. General Characters. — A smaller species, similar in general colora- tion to T.flampcs above described, but brighter ochraceous, feet white, tail longer than head and body, pterygoids narrowed basally, reaching the level of the hinder edge of the palate. Description. — Top of head, nape, and median dorsal region the usual mixture of black and pale ochraceous buff, becoming clearer ochraceous buff on the sides where there is but slight admixture of black hairs. Compared with fiavipcs the ochraceous tint is brighter, but not so bright as in momhasae in which the head and nape are nearly clear, instead of being largely mixed with black. Fore and hind feet covered with short white hair. Area between the eye and ear paler, lower border of eye black. Ventral surface of head and body pure white to the roots of the hairs. Upper surface of tail thinly clothed with coarse, short, black hairs, not appreciably longer at the tip; 334 bulletin: museum of comparative zoology. lower surface covered with short ochraceous buff hairs slightly paler on the distal half, and without admixture of black. Ears externally covered with black hairs, and minutely bordered with whitish. Tail slightly longer than head and body, about 112%; hind foot shorter and stouter than in T. momhasac. Skull. — The skull is that of a fully adult animal with teeth con- siderably worn. The anterior palatal vacuities do not extend quite to the level of the posterior edge of the second molar, and the posterior are present as rounded foramina larger than in T. robusta, beyond which the lateral extension of the palatal is very much more reduced. The pterygoids do not extend be- yond the posterior median edge of the palate and are not expanded proximally. The nasal portion is relatively shorter than in T. robusta. Measurements. — The type meas- ured in the flesh : — head and body 141 mm., tail 158, hind foot 34, ear from meatus 20. Skull : — oc- cipitonasal length 38.5, basal length 32.5, palatal length 20, nasals 15.4, zygomatic breadth 20.4, incisive foramina 6.6, audital bulla 10 X 6.5, upper molar series (alveoli) 7, lower molar series (alveoli) 6.3, upper diastema 10.1. Remarks. — This small Tatera of the Blue Nile does not seem to resemble any of its geographically near allies. From the Abyssinian T. shoana it differs in its untufted tail, smaller size, and proportions. Compared with T. mombasae it is distinguishable at a glance by its shorter hind foot, less clear ochraceous coloring, and its tail which is proportionately shorter, coarser-haired, and differently colored. The t^'pe was brought in by the natives at Fazogli near the Abys- sinian border, and was said to have been caught in the tall grass of the alluvial plains. With it were its four young, about one third grown, which differ in color from the adult in being much darker above owing to the predominance of long black hairs; the sides are only slightly tinged with pale ochraceous buff, and the tail below is clear white to the tip, not ochraceous buff as in the adult. The discovery of these two additional species is of the greatest interest, as hitherto Tatera Fig. 3.- X 3. Tatera soror, palatal region. ALLEN: AL\.MMALS FROM THE BLUE NILE VALLEY. 335 robusfa, a tufted-tailed species, was the only member of the genus kno^^■n from this part of the Nile Valley. Graphiurus orobinus (Wagner). Pigmy Dormouse. Myoxus orobinus Wagner, Abh. Kon. Baier. akad. wnss., 184:3, p. 1-19. There is much doubt as to the identity of Wagner's M. orobinus, the tN-pe locality of which is Sennar. The original description is too brief to be of much avail, nor does Reuvens in his re\new of 1890, shed further light on the subject. The length of the body ( = head and body) is given as 4"2'" or about 107 mm. We obtained five dormice on the Blue Nile, at El Garef and ^Magangani in traps set at the foot of thorn trees in scattered groves with vines and undergrowth. As no other species was met \\'ith, it may be that these represent orobinus though the largest is smaller than Wagner's measurement indicates. They are of the group to which G. parvus belongs, but rather pallid, — a broAMiish gray above, slightly clearer on the shoulders, black eye- ring nearly obsolete, tail pale drab; below whitish, with a tinge of buff. The gray bases of the hairs show through on the abdomen. The tail is not white-fringed. The measurements of two adult fe- males (M. C. Z. 14,483, 14,486) are: —head and body 83, 88; tail 75, 71; hind foot 15, 17; ear from meatus 13.5, 12.5; greatest length of skull 25. It is not unlikely that Wagner's specimen was one of the larger b^o^^'ner group of dormice, and that ours is an undescribed race of the smaller group. DoUman's Graphiurus butleri seems to be a larger species ; it was described from Jebel Ahmed Agar, on the \Miite Nile below Fashoda. EUXERUS ERYTHROPUS LEUCOUAIBRINUS (Ruppell). Side-striped Ground Squirrel. Scuirus leucoumbrinus Ruppell, Neue wirbelth. fauna Abyssinien. Saugeth., 1835, p. 38. We first saw this Squirrel between Sennar and Singa, and it was subsequently met with all along the Blue Nile to Fazogli where we obtained a young one, not more than a third g^o^^'n, in late January. Heuglin states that these animals appear in early forenoon and late 336 bulletin: museum of comparative zoology. afternoon foraging on the ground for food, but we found them about during the hottest hours of the day, running from clump to clump of scattered bushes or herbs, often stopping motionless to look about, and frequently making considerable journeys across open ground. Their holes were almost always found to have several openings close together, whether separate burrows or a common burrow was not ascertained. It was noticeable that the Squirrels were confined al- most wholly to sandy soil, and were practically absent from the hard and sun-cracked " cotton soil." No doubt the latter is of too sticky a consistency in the wet season and so unsuitable for burrowing. Relatively fewer were seen on the Binder than on the Blue Nile. In contrast to the ground squirrels of the genus Xerus seen in British East Africa, this species when running away in alarm or otherwise does not erect its tail at right angles to the body, but trails it inertly behind. Paraxerus sp. Bush Squirrel. This is an extremely rare Squirrel in the Blue Nile valley and seems to occur sparingly near the eastern portion along the Abyssinian border. We met with it but twice and unfortunately failed to secure specimens. A pair was seen in a leafy thorn tree a few miles from Fazogli and on Gebel Fazogli a single one feeding among the branches of a white- barked fig tree with thick green leaves, whose small berry-like fruits are eagerly eaten by many species of birds and by the fruit bats. Felis leg roosevelti Heller. Abyssinian Lion. Felis leo roosevelti Heller, Smithsonian misc. coll., 1913, 61, no, 19, p. 2. Lions are now rare on the Blue Nile. Indeed, the only place where we learned of them was at Omdurman, a small native village above Karkoj, where Dr. Phillips heard one. It was at this same place that I. C. Johnson in 1901, killed a lion; farther up at Soleil, he shot two others, and found more on the southerly bank of the river opposite Bados. Probably they have somewhat decreased in the twelve years intervening for we did not learn of their presence except at Omdurman. Possibly, also, there are more in this region during the rainy season. ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 337 On the upper Binder River there are still a fair number of Lions, following the herds of antelope. They seem to kill a good many GirafiFe as well. Shortly above Khor Galegu we once came out at midday upon the open dry river bed, and discovered an old lioness stalking a fine Waterbuck. She had crept unseen almost within striking distance by taking advantage of the slight inequalities of the sandy ground. One of her full grown cubs was only a short distance away near some sheltering bushes and nearer at hand were three others. The Lions sighted us and made off, leaving the buck unmo- lested. No doubt Lions are decreasing throughout this district. > Heller, in describing this race, refers to it a Sudanese specimen in the collection of the U. S. National ^Museum. The salient characters are the great breadth of the skull in combination with the small cheek- teeth. Felis pardus Linne. Leopard. Felis pardus Linne, Syst. nat., cd. 10, 1758, 1, p. 41. We obtained no specimen of the Leopard, and are unable to assign the proper subspecific title to those observed. Leopards are not rare in the Blue Nile valley and on the Binder, and their tracks were occa- sionally seen in the dusty trails. At Fazogli they were said to be present on the rocky hills, whence they sometimes descended by night to prowl about the native villages. No doubt they feed largely on the Hyraxes that live among the rocks. On the Binder, Br. Phillips came upon a company of monkeys scolding a Leopard among some thick bushes, but it bounded away at his approach. From its boldness and ciuming as well as its ability to conceal itself in little cover, this will probably be about the last of the big cats to be driven from the country by the spread of settlements. Felis capensis phillipsi, subsp. nov. Phillips's Serval. Type. — Adult male, skin and skeleton, 14,908 M. C. Z., from El Garef, Blue Nile, Sudan. January 10, 1913; Br. John C. Phillips, collector. General Characters. — A rather pale, short-tailed form, in which the body stripes are completely reduced to small spots. 338 bulletin: museum of comparative zoology, Bescriytion. — General color of the dorsal surfaces "buff" (of Ridgway, 1886), a shade deeper along the middorsal line. On the forehead, cheeks, and feet are a few small black spots as usual in the servals. The nape is marked by the usual two pairs of black stripes, the outer of which is the broader (about 10 mm.) and runs from the inner base of the ear for about four fifths of the length of the neck beyond which point it breaks into a series of black spots. The inner pair is similar but one half as broad. On the shoulders all the stripes commonly present in the usual served pattern are broken into elongated spots, the largest of which are some 50 mm. long by 10 wide. From the shoulders to the rump the body is uniformly spotted, without any semblance of a stripe posterior to the shoulders. These spots are arranged in more or less definite longitudinal rows, some fifteen in number at the middle region, and average about 10 to 15 mm. in diameter. The ear, as usual, is black on the terminal half with a whitish cross stripe which in the type extends practically across to the inner border. On the fore legs, a black band crosses the dorsal side below the elbow; while on their ventral surface there are two broad black bands. The ventral surface of the body and inside of the legs, chin, and upper throat are white except for the black markings. The lower throat is pale buff. A narrow black band crosses the throat between the angles of the jaws and a second about half the length of the throat. The tail is colored buff with seven black rings, the more basal of which are not quite complete ventrally; the tip is included within the seventh ring. Measurements. — The measurements of the fresh specimen are: — head and body 792 mm., tail 290, hind foot 185, ear from meatus 90. Weight 21 pounds. The tail seems unusually short in this race, about 36% of the head-and-body length, against 46% in hindci and 43% in kempi its nearest neighbors geographically. Skull. — Basal length 102 mm., palatal length 46, zygomatic breadth 78.5, interorbital constric- tion 22, mastoid width 46, upper cheek teeth (front of canine to back of molar) 38, lower cheek teeth (front of canine to back of sectorial) 42, width outside upper molars 45.5. Remarks. — ^NvonghioYi (1910, p. 205) has shown that the name Felis serval, based on an Asiatic cat, if not unidentifiable, is at least untenable for an African species. He proposes to ignore the name in a technical sense, and adopts in its stead Felis capensis of Forster (1781) based on the serval of the Cape of Good Hope. He recognizes as valid races, F. c. galeopardus of Senegal and F. c. togoensis of Togoland and describes three new forms:— F. c. hindei, type locality, Machakos, ALLEN: MAMMALS FROM THE BLUE NILE VALLEY. 339 British East Africa, F. c. IccmpI, type locality, Kirui, Elgon, and F. c. beirac, type locality, Beira, Portuguese East Africa. I have been able to make direct comparison with specimens in the Museum represent- ing the races galcopardus and hindei but from a study of these and of Wroughton's descriptions it is clear that the nerval of the dry flat country of the Blue Nile Valley is distinct from them all. It appears to be very much paler than kempi of the Elgon district and hindei of British East Africa, and represents to an extreme degree the reduction of the striped pattern to one entirely of spots on the body. The shortness of the tail and its color-pattern are also noteworthy. It is a pleasure to associate this fine cat with the name of Dr. John C. PhiUips, to whose enthusiasm and generosity the present collection is due. The type specimen was trapped by him in a scattered growth of thorn trees on the outskirts of the native village of El Garef. Lynx caracal nubica (Fitzinger). Sudan Caracal. Caracal nuhicus Fitzinger, Sitzb. Kon. akad. wiss. Wien, math.-nat. cl., 1869, 60, pt. 1, p. 20.5. The Caracal is apparently uncommon in the region traversed. There are specimens living in the Zoological Gardens at Gizeh, that were caught on the Blue Nile, and we trapped an immature specimen at Magangani a few miles below Roseires. It had come to the bait in the late afternoon within a short distance of our camp. MUNGOS ALBICAUDUS ALBESCENS (Gcoffroy). White-tailed Mongoose. Herpestes albescens Geoffrey, Rev. et mag. zooL, 1839, p. 16. This was the only ^Mongoose we met with. Specimens were ob- tained on the Blue Nile and on the Binder River. It seemed to be one of the commonest of the smaller Carnivora. Although the smaller species of mongoose are active by day, this species apparently is nocturnal. At a camp a few miles above Roseires, one came to a trap within a few yards of the tent in the early evening, doubtless the same animal that succeeded in stealing the bait from a trap even nearer the preceding evening. At Magangani, on visiting the traps 340 bulletin: museum of comparative zoology. shortly after dawn, one was found already dead ha\Tng been bitten through the neck by a leopard whose tracks were seen in the path. On another occasion we startled one in the early afternoon, that had been ensconced in the hollow under the roots of a fallen tree, no doubt asleep. Genetta abyssinica (Riippell). Abyssinian Civet-cat. Viverra abyssinica Riippell, Xeue wirbelth. fauna Abyssinien. Saugeth. 1835, p. 33, pi. 11. Along the Blue Nile and the Dinder River this seemed to be a com- mon species. Specimens were trapped at Bados and Magangani on the Blue Nile and at the latter spot Dr. Phillips shot one that was clambering up the trunk of a large baobad tree in the full sunlight of noon. At Bados, one was caught in a trap and found next morning partly eaten by a large cat, apparently a Caracal, that bounded off in the dusk when surprised. Curiously, we did not succeed in trapping any in the more northern part of our journey between Sennar and Bados, where perhaps they are less common. The extraordinary amount of color variation in this group renders the division into races a matter of much uncertainty. Professor Matschie (1902) in his re\-iew of the civet-cats, was able to examine some 240 skins in the Berlin Zoological ^Museum, and recognized no less than thirtv-three forms, all of which mav be considered races of two species, the one ^\'ith a longer-haired, the other with a shorter- haired tail. In the latter group belong the specimens obtained by the Phillips Expedition. Although the propriety of recognizing so many local races may be questioned and the value of certain of the characters considered distinctive is yet to be showTi, the four skins preserved do agree in having the light tail annulations much wider than the dark, and the feet practically of the same hght gray on both the superior and the inferior surfaces, marks which Matschie finds distinctive of the civet-cats of the Red Sea coast {G. schraderi from jNIassawa) and the present species, described by Riippell from between Kordofan and Gondar in Abyssinia. As these specimens are practi- cally topotypes of abyssinica, a brief statement of the variation in color is of interest. This is mainly a matter of the relative amounts of black, rusty, and buff in the pattern, and the degree to which the rows of spots coalesce to form stripes. In two specimens, the ground color ALLEN: MAMMALS FROM TUE BLUE NILE VALLEY. 341 is uniformly pale buff; the two dorsal rows of spots on each side are much larger in one than in the other and prevailingly rusty in color. The median stripe is likewise more rusty than black. In the other two skins, the ground color is clearer gray, the dorsal rows of spots in one case rusty in the other more black than rusty, and the dorsal stripe black. In all, the two terminal light rings on the tail are in- complete dorsally owing to the median black portion connecting the three last dark rings. The stripes and rows of spots vary even on opposite sides of the body. The spots in the row nearest the midline on each side show a marked tendency to run together into a stripe over the hips. The outer stripe from the nape to the shoulder, in one individual is broken into a series of elongated spots. The pale annulations of the tail in all are white ventrally shading rather abruptly into buff on the dorsal side. At the ankle, posteriorly the dark spot is rather ill defined and restricted. All the three males in the series are more buff than the single female, but the latter is practically identical w^th one of the males. Both are from Magangani, some ten miles below Roseires on the Blue Nile, whereas the two buffer speci- mens, with rusty spots and median stripe are from the Dinder River at Kuka and Ereif el Dik respectively. Hyaena hienomelas Matschie. Nubian Striped Hyaena. Hyaena hienomelas Matschie, Sitzb. Ges. naturf. freunde Berlin, 1900, p. 53. The Striped Hyaena of the Atbara and neighboring region is con- sidered distinct by Matschie in his revision of the species. He calls it H. hienomelas, and quotes Latreille (Sonnini's Suites de Buffon, 27, p. 25) as the- authority. Latreille, however, did not give a Latin designation to this species, but refers to a specimen in the Paris Museum as having been called by Lacepede ckieti hienomelas. He further quotes Bruce's account of its habits in the Sudan. The Latin name must then apparently be credited to Matschie. We trapped a specimen at Magangani and several times heard them about our camps along the Blue Nile. What I took to be their cry is differ- ent from that of the Spotted Hyaena, having a more musical quality with a rising then a falling inflection. 342 bulletin: museum of comparative zoology. IcTONYX erythraea de Win ton. Red Sea Striped Weasel. Ictonyx erythraea (sic) de Winton, Ann. mag. nat. hist., 1898, ser. 7, 1, p. 248. The type locality of this species is Suakin on the Red Sea coast, and its describer considers that a specimen from Somaliland represents the same form. A male collected by Mr. A. L. Butler at Roseires on the Blue Nile is also referred to this species by Wroughton (1911, p. 459). Two specimens were taken by our expedition — at Gabardi and El Garef respectively, localities between Singa and Roseires. Erinaceus albiventris pruneri Wagner. White-bellied Hedgehog. Erinaceus pruneri Wagner, Schreber's Saugeth. suppL, 1841, 2, p. 23. Although apparently not uncommon, Hedgehogs were hard to ob- tain. A live one was brought in by the natives at Fazogli who said they occasionally came upon them, or found them hidden in hollow logs or tree trunks. The dried spiny portion of the skin is sometimes found, as if left by some animal that had eaten the rest. This species belongs to the group for which Pomel in 1848 proposed the generic name Atelerix, type species pruneri. Fitzinger in 1867 gave the name Peroechinus to the same group of small hedgehogs that lack the first hind toe. In the specimen from Fazogli the toes are very short, hardly separate from the pad. The claw of posterior digit 2 is largest, curved, and flattened. The remaining claws of the hind foot are successively smaller, that of the third digit rather flat- tened, those of the fourth and fifth compressed laterally. The face from nose to between the eyes is thinly covered with short dark brown hairs, and in life the skin is blackish. The hair of the forehead, cheeks, and ventral surfaces is dull white, mixed on the ears, legs, and tail with brown. This coloration separates it from the Senegambian albi- ventris of which it is made a subspecies by Anderson and de Winton in their Mammalia of Egypt. The spines are blackish, with white tips, and a few along the sides are white throughout. We found nothing of the species senaarensis described from Sennar Province, It belongs to the group of larger species with five well- developed claws on the hind foot. According to Anderson and ALLEN: AIAMMALS FROM THE BLUE NILE VALLEY. 343 de Winton the locality "Senaar" is very doubtful, and it is probable that the name is a synonym of E. aethiopicus of lower Egypt. Crocidura sericea (Sundevall). Silky Shrew. Sorex sericeus Sundevall, ex Hedenborg MS., Kongl. Svenska vet.-acad. Handl. for 1842, 1S43, p. 173. In his essay on the genus Sorex above cited, Sundevall in 1S43 de- scribed three new Crociduras from Sennar and the ^Yhite Nile on the basis of specimens sent by Hedenborg, the Swedish traveller. The first of these, Sorex (= Crocidura) hedcnborgianiis, is characterized as a rather large species, head and body 140 mm., tail 52, skull 31 mm. long, of a uniform chocolate-brown above and below. We found nothing of this animal. The second species S. (= C.) fidmster is said to be pale gra^^sh bro\\ii above, ashy white below, the tail about half the length of head and body. The single specimen came from the White Nile, and measured: — head and body 90 mm., tail 44, skull 21 mm. long and 5 mm. between the orbits. The third species