® 4 t od " fi . 2 yon bay A mee 7 i J val ii ¥ 4 sii — an se sogy a a ee ee en eee ae i vi ia a | it bee a al eat | a 4 F : \ if i ti) ny, ie we iy gt |, | SMITHSONTAN MISCELLANEOUS COLLECTIONS VOL WwW “EVERY MAN IS A VALUABLE MEMBER OF SOCIETY WHO, BY HIS OBSERVATIONS, RESEARCHES, AND EXPERIMENTS, PROCURES KNOWLEDGE FOR MEN ””__sMITHSON (Pus.ication 4023) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION 1950 The Lord Wattimore Press BALTIMORE, MD., U. S. As ADVERTISEMENT The Smithsonian Miscellaneous Collections series contains, since the suspension in 1916 of the Smithsonian Contributions to Knowledge, all the publications issued directly by the Institution except the Annual Report and occasional publications of a special nature. As the name of the series implies, its scope is not limited, and the volumes thus far issued relate to nearly every branch of science. Papers in the fields of biology, geology, anthropology, and astrophysics have predominated. A. WETMORE, Secretary of the Smithsonian Institution. (ili) PEM TL oo Gh ae (irae | ai bi: vg ia gdal » ad aie "hg ie By ag ra 0 Rayon 7 ae et, al ae ¢ eis ua ee et Omar mr eh ee | oe” ae a A ‘i vial ie i iken' ni : Sih nei ae cy : fs sis oad ay Pyae PORT sia eA i \ B ( “i NS) 10. 12. 13. 14. LS. CONTENTS . FRIEDMANN, Herpert. Mirandolle’s forest falcon. 4 pp., 2 pls. Sept. 22, 1948. (Publ. 3948.) WepeL, Watpo R. Prehistory and the Missouri Valley develop- ment program. Summary report on the Missouri River Basin Archeological Survey in 1947. 52 pp., 8 pls. Nov. 23, 1948. (Publ. 3950.) Knicut, J. Brookes. Further new Cambrian bellerophont gas- tropods. 6 pp., I pl. Dec. 24, 1948. (Publ. 3951.) . Cuapin, Epwarp A. Type material of the species of clerid beetles described by Charles Schaeffer. 12 pp. Apr. 5, 1949. (Publ. 3966.) Asgot, C. G. 1948-1949 report on the 27.0074-day cycle in Washington precipitation. 2 pp. Mar. 8, 1949. (Publ. 3980.) Azspot, C. G. A prediction of Washington temperature 1948 (made January 1948). 6 pp., 1 fig. Mar. 8, 1949. (Publ. 3982.) Aspot, C. G. Montezuma solar-constant values and their periodic solar variations. 13 pp., 3 figs. Apr. 19, 1949. (Publ. 3981.) Raw, Frank. Some stages in the evolution of the nervous system and the fore-gut of the polychaet. 35 pp., 5 figs. Aug. 4, 1949. (Publ. 3983.) FRIEDMANN, Hersert. A new heron and a new owl from Venezuela. 3 pp. July 21, 1949. (Publ. 3985.) HILDEBRAND, SAMUEL F. A collection of fishes from Talara, Pert. 36 pp., 9 figs. Aug. 18, 1949. (Publ. 3986.) . GLEN, Rospert. Larvae of the elaterid beetles of the tribe Lepturoidini (Coleoptera: Elateridae). 246 pp., 4o figs. Apr. 19, 1950. (Publ. 3987.) AvcpricH, L. B. Note on Fowle’s spectroscopic method for the determination of aqueous vapor in the atmosphere. 6 pp., I pl., 2 figs. Sept. 20, 1949. (Publ. 3989.) Assot, C. G. Short periodic solar variations and the tempera- tures of Washington and New York. 8 pp., 2 figs. Oct. 4, 1949. (Publ. 3990.) AcpricH, L. B. The Abbot silver-disk pyrheliometer. 11 pp., {pl 1 hie. Dec: & 1949. (Publ.3601.) Fenton, Witt1AmM N. The roll call of the Iroquois chiefs. A study of a mnemonic cane from the Six Nations Reserve. 73pp. ta. pls, 3 est ev. 10.1950, (CEubl, 39605:) (v) vi SMITHSONIAN MISCELLANEOUS COLLECTIONS 16. FRIEDMANN, Herpert. The forms of the black hawk-eagle. 4 pp., I pl. Feb. 28, 1950. (Publ. 4013.) 17. Appot, C. G. Periodic influences on Washington and New York weather of 1949 and 1950. 8 pp., 3 figs. Mar. 22, 1950. (Publ. 4015.) 18. Grocx, Watpo S. Tree growth and rainfall—a study of corre- lation and methods. 47 pp., 7 figs. Oct. 25, 1950. (Publ. 4016.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 1 MIRANDOLLE’S FOREST FALCON (With Two PLaAtTEs) BY HERBERT FRIEDMANN Curator, Division of Birds, U. S. National Museum . (Pustication 3948) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION SEPTEMBER 22, 1948 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 1 MIRANDOLLE 'S FOREST FALCON (Wits Two PLatEs) BY. HERBERT FRIEDMANN Curator, Division of Birds, U. S. National Museum ry F 17 ot = o 4eVs5HS VCore HINTON (PuBLicaTIon 3948) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION SEPTEMBER 22, 1948 The Lord Baltimore Press BALTIMORE, MD., U. 8. A. MIRANDOLLE’S FOREST FALCON By HERBERT FRIEDMANN Curator, Division of Birds U. S. National Museum (WitH Two PLatEs) Mirandolle’s forest falcon, Micrastur mirandollei (Schlegel), is one of the least known of American hawks in spite of a very extensive range, extending from Costa Rica to Bolivia, and it has long been a rare bird in museum collections. So little material was available for study that in 1932 Griscom (Bull. Mus. Comp. Zool., vol. 72, p. 317) was moved to write of Wedel’s eastern Panamanian collection that “a series of five specimens of this very rare Hawk from one locality is unprecedented. . . .”” Since that time additional examples have been taken, and in 1941 Griscom and Greenway (Bull. Mus. Comp. Zool., vol. 81, p. 418) separated the birds of the Caribbean coast of extreme eastern Panama under the name M. m. extimus, as they ob- served that two adults from there differed from a single Brazilian adult and from the description of the type from Surinam in having “much narrower and paler tailbands and in having the white under- parts washed with richer buff. Judging by an immature bird and an intermediate specimen also from eastern Panama, the differences de- scribed above have nothing to do with immaturity... . ” Their com- parative material was admittedly inadequate, but, fortunately, their action proves to be correct. Through the courtesy of the officials of the American Museum of Natural History, Carnegie Museum, Chicago Natural History Mu- seum, Museum of Comparative Zoology, and the Academy of Nat- ural Sciences of Philadelphia, I have been able to add to the limited material stored in Washington, and have brought together by far the largest and most representative series of this species ever assembled, 26 specimens in all. These birds, young and adult of both sexes, clearly substantiate the validity of extimus, as the accompanying fig- ures (pls. I, 2) show, but also indicate that its range is much more extensive than hitherto thought. It is because of the rarity of this hawk that I have felt it worth while to illustrate the races in this paper, as otherwise it would take an equal amount of borrowing for anyone SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 1 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS nt else to assemble an adequate corpus of data to elucidate them. The illustrations show representative samples of both races, including in each case immature as well as adult birds, and males as well as fe- males. To show more examples in each picture would have reduced their scale too much without in any way adding to the demonstration of the distinctness of their racial characters. The races and their ranges are as follows: Micrastur mirandollei extimus is characterized by having the pale bands on the dorsal surface of the rectrices narrow, white or almost white with a medial dusky band. This is true of young and old birds alike. The series studied does not uphold the supposed character of a richer buffy wash on the white underparts. This race ranges from Costa Rica (Talamanca) and Panama (scattered records from the Caribbean slope, one from the Pacific slope, Permé, Obaldia, Panama Railway Line), south to the Chocd area of western Colombia (Rio Jurado, Rio Baudd, Rio Yarubida, where a series of 10 specimens was collected by von Sneidern). Micrastur mirandollei mirandollei has the pale bands on the dorsal surface of the rectrices somewhat broader and much duskier—grayish brown, becoming whitish on the inner webs of the lateral tail feathers, and, in young birds, the dusky margins of the feathers of the throat and breast paler, more washed out than in corresponding examples of extimus. This form ranges from Colombia east of the eastern Andes (Meta (Quenane) and Caqueta)' to Venezuela (Cerro Yapa- cana, Upper Orinoco; Cerro Duida), and the Amazonian basin of 1 Unfortunately, I have not seen either of these eastern Colombian specimens, and Nicéforo Maria’s example from Meta is figured (Caldasia, vol. 4, No. 19, p. 321, May 30, 1947) from the underside and consequently does not show the dorsal appearance of the tail bands. It is, however, a young bird, and, as far as may be judged from the photograph, has fairly pale edges to the pectoral and upper abdominal feathers, which would suggest its being of the nominate race. This is not clarified or too well borne out by the accompanying descrip- tion, wherein it is said that the tail feathers are tipped with white, but have the other bars grayish brown, with white edges, which might even seem to suggest extimus! I understand that during the recent uprising in Bogota, Brother Nicéforo’s collections were lost in the destruction of the building that housed them. No eastern Colombian examples are known in any collection outside of Colombia. Fortunately, in response to my inquiry, Dr. Armando Dugand has supplied the necessary information to enable me to place the Caqueta bird (and hence, by inference, the Meta example) as typical mirandollei. He writes me that the specimen from Morelia, Caqueta, has the two central rectrices with “three bars plus one terminal; the bars are grayish brown, only very slightly paler than the dark brown of the feathers . . . the other rectrices show white bars only on the inner webs. . . .” NO. I MIRANDOLLE'S FOREST FALCON—-FRIEDMANN 3 Brazil (Rio Jurua, Villa Braga, Santarém, Tapajoz River, Beireio, Rio Majary, Rio Acora, Rio Jamunda, Sao Gabriel, Igarapa Aniba, Piquiatuba, Manacapuru, and Allianca, Para), southeast to Espirito Santo (Rio Doce), and northeast to British Guiana (Carimang River, Ituribisi River, Courantyne, Bartica, Arawai River), Dutch Guiana, and French Guiana (Cayenne), and south to eastern Perit (Chy- avetas, Yurimaguas) and to Bolivia (Santa Ana, near La Paz, on the Rio Corioco). The nominate race averages larger than the northern form as the following measurements show, but the overlap is too great for size to be used as a diagnostic character. Micrastur mirandollei mirandollet: 11 males (including 5 from the literature), wing 220-241 (230.5) ; tail 179-202 (189.1) ; culmen from cere 20-23 (20.8) ; tarsus 70-76.2 (75.5); middle toe without claw 35.7-37.2 (36.7 mm.) ; 7 females, wing 221-248 (233.0) ; tail 180-208 (190.7) ; culmen from cere 19.9-22.4 (20.9) ; tarsus 73-86 (77.9) ; middle toe without claw 34.2-38.8 (36.2 mm.). In this connection it may be mentioned that Hellmayr (Nov. Zool., vol. 17, p. 409, 1910) gives the measurements of an old adult female from Allianca, Brazil, as wing 265, tail 200, culmen 24, and tarsus 55 mm. The wing is much longer and the tarsus much shorter than in any birds I have examined. Micrastur mirandollei extimus: 7 males, wing 215-237 (221.3) ; tail 182-200 (191.5) ; culmen from cere 20-23.3 (21) ; tarsus 74.1-81 (77.6) ; middle toe without claw 32.5-38 (35.1 mm.) ; 7 females, wing 220-228 (224.3); tail 187-195 (191); culmen from cere 20.2-22.4 (21.4) ; tarsus 78-84 (80.1) ; middle toe without claw 35.1-38 (36.1 mm.). It appears from these figures that this is one of the relatively few hawks in which there is little or no difference in size between the sexes. A detailed description of its plumages, given below, completes our present knowledge of this rare forest falcon. Micrastur nurandollet mirandollei: ApuLtT (sexes alike in colora- tion) : Forehead, crown, occiput, nape, scapulars, interscapulars, back, rump, upper tail coverts, upper wing coverts, and remiges deep neu- tral gray with a slate tinge, the top of head somewhat darker; the inner webs of the remiges white for their basal two-thirds or more, and crossed by four fairly broad dark mouse-gray bands and somewhat mottled with deep mouse gray on the more distal portion of the whitish area; the fifth (from the outside) primary the longest, the first (outermost) the shortest; tail chaetura black tipped with white or buffy white and crossed by three fairly broad grayish-brown bands, these bands becoming whitish on the inner webs of the lateral 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III feathers ; lores and orbital area bare, a slight line below the eye, the cheeks, and the auriculars like the crown; the posterior auriculars shading to buffy white; chin, throat, breast, abdomen, sides, flanks, thighs, and under tail coverts whitish to pale light pinkish buff, the feathers, especially of the breast and sides, often, but not always, with very fine dusky shaft lines; under wing coverts whitish to pinkish buff, the lower ones mixed or banded with deep neutral gray; iris yellowish brown; cere and base of both mandibles yellowish, the bill otherwise horn black; tarsi and toes bright yellow ; claws black. IMMATURE: Similar to the adult, but with a brownish slate tinge on the upperparts; the pale tail bands washed with pale fulvous; the entire underparts with a buffy wash; the feathers of the breast and sides with prominent dusky shaft streaks ; bill blackish except basally where it is yellow. JuvenaL: Similar to adult, but upperparts dull fuscous with a faint slate wash; underparts whitish with a buffy wash on the throat, breast, sides, flanks, and upper abdomen, but not on chin or vent; each feather of the throat, breast, sides, flanks, and upper abdomen with a broad border of fulvous drab to hair brown, producing a con- spicuously scalloped appearance ; bill yellow, slightly dusky along the proximal portion of the culmen. NATAL DOWN: Unknown. Micrastur mirandollei extimus: Aputt: Like that of the nominate race except for the color of the tail bands, as shown in the illustra- tions and mentioned above. (No specimen from the range of e+- timus shows any approach to mirandollei, but one adult female of the latter form, from Rio Jamunda, Brazil (Amer. Mus. Nat. Hist. 283244), and one from Cerro Duida, Venezuela (Amer. Mus. Nat. Hist. 272322), show some resemblance to extimus in this respect. However, 2 specimens out of II is not a serious exception to the gen- eral picture, and of these 2, the Cerro Duida bird is less aberrant than the one from Rio Jamunda. ) IMMATURE: Similar to that of typical mirandollei. JuvENAL: Similar to that of the nominate race but with the broad edges of the feathers of the throat, breast, sides, flanks, and upper abdomen averaging darker—dark hair brown. NATAL DowN: Unknown. Specimens examined.—M. m. mirandollei, 11: Venezuela 2, Brazil 6, British Guiana 1, French Guiana 1, Bolivia 1. M. m. extimus, I te Costa Rica 1, Panama 4, western Colombia 9, without locality 1. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE Sela iNO Pie MICRASTUR MIRANDOLLEI MIRANDOLLE!I (SCHLEGEL) Upper, dorsal view; lower, ventral view of same individuals (1 immature, 5 adult) showing lack of correlation between age and the character of the dorsal tail band. The specimens figured are from Venezuela, Brazil, British Guiana, French Guiana, and Bolivia. SMITHSONIAN MISCELLANEOUS COLLECTIONS VO) NOS SP ee MICRASTUR MIRANDOLLE!I EXTIMUS GRISCOM AND GREENWAY Upper, dorsal view; lower, ventral view of same individuals (4 juvenal, 2 adult ) showing that age is not a factor in the character of the dorsal tail eee The specimens figured are from Costa Rica, Panama, and western ~OLOMmMD1a, a a re ee ‘SMITHSONIAN MISCELLANEOUS COLLECTIONS Ler VOLUME 111, NUMBER 2 PREHISTORY AND THE MISSOURI VALLEY DEVELOPMENT PROGRAM | SUMMARY REPORT ON THE MISSOURI RIVER BASIN ARCHEOLOGICAL SURVEY IN 1947 (Wirth Etcut Prates) AL BY WALDO R. WEDEL Associate Curator, Division of Archeology, U. S. National Museum (PUBLICATION 3950) "oO. GrEy OF WASHINGTON __. » PUBLISHED BY THE SMITHSONIAN INSTITUTION WEES sh 7 NOVEMBER 23, 1948 wh SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 2 PREHISTORY AND THE MISSOURI VALLEY DEVELOPMENT PROGRAM SUMMARY REPORT ON THE MISSOURI RIVER BASIN ARCHEOLOGICAL SURVEY IN 1947 (With Eicut PLAtEs) BY. WALDO R. WEDEL Associate Curator, Division of Archeology, U. S. National Museum SHON 007 SHINTO 11 @Qeo® (PusticaTion 3950) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION NOVEMBER 23, 1948 The Lord Baltimore Mress BALTIMORE, MD., U. 8. As PREEPISLORY AND THE MISSOURL VALEEY DEVELOPMENT PROGRAM SUMMARY REPORT ON THE MISSOURI RIVER BASIN ARCHEOLOGICAL SURVEY IN 1947 By WALDO R. WEDEL Associate Curator, Division of Archeology, U. S. National Museum (WitH Eicut PLATEs) INTRODUCTION This report summarizes the field and laboratory activities in arche- ology and paleontology by the Missouri River Basin Survey during the calendar year 1947. It is not a complete or final statement of accomplishments during the year, nor does it undertake to set forth the opinions of the various staff members who have been directly responsible for the field and laboratory researches, and whose findings constitute much of the basic information on which this summary is based. Essentially, it is a report of progress as of December 31, 1947, at the end of the first 18 months in a scientific salvage program linked to “the most comprehensive and far-reaching river basin de- velopment plan ever undertaken in America’’—the harnessing of the Missouri River and its tributaries. The general background, organization, and basic objectives of the Missouri River Basin Survey have been adequately set forth else- where and need not be detailed again here.t The project represents but one regional phase of the River Basin Surveys, a nation-wide archeological and paleontological scientific salvage program under the direction of Dr. F. H. H. Roberts, Jr., for the Smithsonian Institution. This program is based directly on a memorandum of understanding formulated in 1945 between the Institution and the National Park Service, and indirectly on a series of interbureau agreements between the Park Service, the Bureau of Reclamation, and the Corps of Engineers. Its purpose, briefly, is to locate, record, 1 See Smithsonian Misc. Coll., vol. 107, No. 6, Apr. 23, 1947; and Amer. Antiq., yol. 12, No. 4, pp. 209-225, April 1947. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 2 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III and evaluate the archeological and paleontological resources that will be affected by the many Federal water-control projects planned or under construction by the Bureau of Reclamation, Department of the Interior, and the Corps of Engineers, Department of the Army ; to transmit this information to the National Park Service; and to recommend, where necessary, the procedures needed for recovery of as much as possible of the scientific information which would other- wise be lost. As excavation at key sites supersedes the survey and test digging which alone have so far been possible, it will be the Smith- sonian’s responsibility to direct the Federal phases of this work as well. Funds to support the operations thus far have come from the Bureau of Reclamation through the National Park Service to the Smithsonian. For the most part, they have been for survey only. During fiscal year 1948, limited excavation funds were made available for work at Angostura, S. Dak., Boysen, Wyo., and Heart Butte, N. Dak. ; of these units, only Boysen was visited for limited excava- tions during the calendar year 1947. In the Missouri Basin, as elsewhere, American archeology will long be indebted to the governmental agencies whose efforts and financial support are making possible the salvage operations. Through the enlightened conservation policy and excellent cooperation of the National Park Service and the Bureau of Reclamation, what threatened to be a program of wholesale destruction of scientific re- sources in the Basin promises, if properly carried through, to be- come one of the most comprehensive archeological and paleontological research projects ever undertaken in this region. It is a pleasure to note here that throughout 1947 the relationships of the Missouri River Basin Survey with other agencies have re- mained, on the whole, cordial and pleasant. Close contact has been maintained at all times with the Missouri River Basin Recreation Survey, Region 2, National Park Service, in which office Chief Rec- reation Planner Guy D. Edwards and Archeologist J. D. Jennings have been particularly helpful. The Bureau of Reclamation, besides financing the work, has freely furnished maps, construction schedules, and other materials and information, as requested. The Corps of Engineers has likewise been generous in providing topographic maps and other information ; in the Omaha District office T. E. Huddleston has been very helpful in the archeological interpretation of aerial photographs of the upper Missouri Valley. Local and regional repre- sentatives of all these agencies, project engineers, and others have been uniformly cooperative. The same may be said in regard to State and other non-Federal agencies, as well as of numerous private indi- NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 3 viduals. The active interest of the archeological profession and of learned societies, as expressed through the Committee for the Recov- ery of Archeological Remains, has also been of very real assistance. The Missouri River Basin Survey was established in July 1946, with headquarters at Lincoln, Nebr. In 1946, and throughout most of 1947, field activities were in the main restricted to preliminary reconnaissance and survey at the most urgent projects, that is, where actual dam construction had been initiated or where preconstruction activities were nearing completion and construction was expected to start in the near future. Some test-pitting and small-scale excavation was undertaken near the close of the 1947 field season, but in general it can be said that the work during the first 18 months constituted the initial phase of the program. By the end of 1947, reconnaissance parties of the River Basin Surveys had visited and partially or completely surveyed 44 Bureau of Reclamation and 6 Corps of Engineers projects in the Missouri River watershed. These were distributed throughout seven States, as follows: Colorado, 3; Kansas, 6; Montana, 4; Nebraska, 16; North Dakota, 10; South Dakota, 5; Wyoming, 6. A total of 508 sites of archeological interest were located and recorded, many of them previously unreported. That this number probably represents only a fraction of those which actually await discovery and recording is indicated by the many new sites found in intensive survey at certain proposed units which had previously been examined only in prelimi- nary fashion. Thus, for example, at Glendo in Wyoming 8 sites were recorded during a 4-day reconnaissance in 1946, but 35 additional locations were hunted out during a 3-week visit in 1947. Other units gave comparable results, thus demonstrating the need for thorough search at any proposed reservoir site before water is impounded. A list of reservoir projects visited from July 1946 to December 1947, with the number of archeological sites so far recorded for each, follows: Bureau of Reclamation: RESERVOIRS SITES RESERVOIRS SITES Tp AIMHeETStINEDis, sc ceee sce 5 o: Brewster, Nebr: s.....--- I A, Jinan AWAOk cao cooboobe 3 tom Bronco Ns Dalk.t sec - 9 3. Angosturd; S: Daki). 0.s% 5 11. Buffalo Creek, Dawson 4, Beaver City, Nebr... o.:<:c. 4 CountysNebry cents I Bevbliewtorsey Ss Walco ace 0 12. Buffalo Creek, Dundy (6), 1xoyarMy, (OOO), Gosodcoocncc 3 (Comming INR Boodooces I 7. Box Butte; Nebr... sos sete I 11g} (Carngoy INGIes Goadbobanads (a) &. boysem Wyo. v.20. ocsee 75 14. Canyon Ferry, Mont..... 33 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, (Pit Bureau of Reclamation (Cont'd) : RESERVOIRS SITES RESERVOIRS SITES M5 @edar woluthyslGans\ oe ee 4 30. Medicine Creek, Nebr.... 20 TO! WeTOSDY.. (Nie WARK ci o.c he cis I 31. Medicine Lake, Mont.... 8 17. Culbertson, Nebr. ....... I 32) Mullens SNebra eeeeerr one 8 19.) Deerneldsss Dakin. +.se ) 33:vNortony (Kans) fost oamenne 3 TOspDesulsacss Nim Malksenm eee 0 34. Oregon Basin, Wyo..... 28 20. Devils Lake, N. Dak..... 5 35.) Pioneer ans. eee I Bree Wickinsonny Nea kettle 3 36. Red Willow, Nebr....... 5 Poy, Malic Noes Wood sopeac 4 37. Rock Creek, Nebr....... I PB wil aleoore IN| Gece sddes 5 39e Rockvale iNebi meee ere 0 Pil, (GIG) WAKOS ohodaauce.c 43 30. Shadehill)'S. Dak.2a5.752 6 25. Heart Butte, N. Dak.... 4 40. Sheyenne, N. Dak........ II 26. Jamestown, N. Dak...... 7 Ar. duibetsalVliont-eeeee eee 53 27 AGNI, 5 WTS Hae eee see I 42. Wilson, Kans, 0.0 »-ee ee 6 Ay iors WAKO ooochooooac I 43: Wray;. Golo. \32,.05- ace 5 29. Lake Solitude, Wyo...... 0) 44. Yellowtail, Mont. ....... 3 Corps of Engineers: Als, IBeNkelaiibl, IN IDEN seo6u0c 10 49. Harlan County, Nebr.... 23 46. Cherry Creek, Colo. .:... 6 50. Kanopelis, Kans. :....2-- 18 A7> Hort Randalls Sa Dalx: «2.03 -— Ash Garcison, Ne Dake. eyes 70 508 The results of paleontological work by the Missouri River Basin Survey during 1947 are summarized elsewhere in this report. PERSONNEL The professional staff throughout the year included six archeolo- gists: Paul L. Cooper, in charge of the field office during the writer’s absences in Washington; Robert B. Cumming, Jr., laboratory super- visor ; Wesley L. Bliss, Marvin F. Kivett, J. J. Bauxar, and Jack T. Hughes. Dr. T. E. White, on leave of absence from the Museum of Comparative Zoology, Harvard University, joined the River Basin Surveys in April, and throughout most of the year he was engaged in paleontological field work in the Missouri River Basin. Full-time laboratory and office personnel included Mrs. Ina May Reagan, advanced from temporary office assistant to clerk-stenog- rapher ; Dean Clark, laboratory assistant; and J. M. Shippee, expert laborer. Drafting; darkroom work; typing of field notes, specimen catalogs, other records and reports ; assembling of the latter ; and some of the routine processing of specimens were carried on with part-time student and other help. Student assistants from several colleges and universities were em- ployed on the summer survey field parties from June to September, as noted in a later section of this report. George Metcalf was added to the roster in October as field and laboratory assistant. NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 5 LABORATORY ACTIVITIES The project field office and laboratory for the Missouri River Basin Survey are located in the Laboratory of Anthropology of the Uni- versity of Nebraska, in the basement of Love Memorial Library. Additional storage space for specimens, equipment, and vehicles has been acquired at the Lincoln Municipal Airport. Selection of Lincoln as the place for the project headquarters has proved generally satis- factory ; and the quarters generously made available by the university have been adequate, in general, for operations on the scale followed up to the present. Laboratory space has also been provided by the Nebraska State Historical Society. During the early part of the year, the organization of the basic laboratory files was completed. These include site survey records, prepared in triplicate and filed for laboratory use by State, county, and reservoir, plus a reserve file. The basic site file at year’s end included 443 site folders, each including a survey sheet, site catalog of artifacts and photographs, and other pertinent data. The reservoir site file consisted of 25 volumes and the reserve file of 17. The map file, which is rapidly being expanded, contains 566 reference maps of various kinds and 269 aerial photographs. Drafting, map-making, and photographic work have gone forward steadily, although reliance has been almost wholly on part-time assistance. Three subbasin location maps, 62 individual reservoir maps, and approximately 700 reproductions of survey maps, sketches, and field-note diagrams have been prepared. Darkroom work includes the processing of 927 negatives representing 210 sites from 7 States; and preparation of approximately 2,500 black-and-white prints to illustrate reports and for other purposes. Kodachrome transparencies to the number of 180 augment the photographic record of accom- plishments to date. By the end of the calendar year, nearly 50,000 specimens had been cleaned, cataloged, and stored. These represent 580 sites in 44 reservoir areas scattered over 7 States. Shortage of laboratory help to conduct preliminary analysis, classification, and recording of speci- mens before they go into storage poses a major problem. Lack of a preparator for the proper handling and restoration, where feasible, of outstanding specimens is another serious handicap. For the designation of archeological sites, and of specimens and photographs therefrom, a relatively simple code system has been adopted by the Missouri River Basin Survey. Site designations are trinomial in character, consisting of symbols for State, county, and site. The State is indicated by the first number, according to the 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III numerical position of the State name in an alphabetical list of the United States ; thus, for example, Kansas is indicated by 14, Nebraska by 25, Wyoming by 48. Counties are designated by a 2-letter abbre- viation; for example, FT for Frontier County, HN for Harlan County, CH for Charles Mix County. The county symbol is followed by another number referring to the specific site within the indicated State and county ; usually, but not necessarily, this number represents the order in which the sites were reported or discovered. Specimens are marked in the laboratory with the appropriate site symbol, fol- lowed by a serial number referring to a particular object or group of closely associated objects. Photographs are similarly designated before filing. This system is an elaboration of that devised in Nebraska during the WPA archeological programs. Up to the present time, very little has been published concerning the archeological survey in the Missouri River Basin. In accord with the memorandum of understanding between the Smithsonian Institu- tion and the National Park Service, however, preliminary mimeo- graphed appraisals of the archeological resources of reservoirs in- vestigated have been furnished to the latter agency. These reports contain site location maps, brief descriptions of materials seen, and an evaluation of the remains located, together with recommendations for further action, if needed. They are distributed chiefly to the con- struction agencies, district and regional officials of the several Federal agencies concerned, and on a selective basis to cooperating institutions and organizations where their particular fields of interest are involved. Thirty-two preliminary reports on archeology were issued during 1947. They include Anchor, Angostura, Box Butte, Boysen, Broncho, Canyon Ferry, Cedar Bluff, Cherry Creek, Crosby, Deerfield, Des Lacs, Devils Lake, Dickinson, Enders, Fort Randall, Glendo, Harlan County, Heart Butte, Jamestown, Kanopolis, Kirwin, Kortes, Lake Solitude, Medicine Creek, Medicine Lake, Oregon Basin, Shadehill and Blue Horse, Sheyenne, Tiber, Wray, and Yellowtail Reservoir areas, and the Lower Platte Subbasin. Included in the last-named were reports on the proposed Amherst, Buffalo Creek, Cairo, Ericson, Mullen, and Rockville Reservoir areas, all in Nebraska. Preparation of preliminary appraisals is well under way for Norton, Beaver City, Red Willow, Culbertson, Rock Creek, Buffalo Creek, Pioneer, and Bonny Reservoir areas in the Republican River Basin; for Wilson Reservoir area in the Smoky Hill Basin; and for Garrison and Bald- hill Reservoirs, in North Dakota. Two preliminary reports on paleontological survey were also issued. One covers Cedar Bluff, Glen Elder, Kanopolis, Kirwin, and Webster NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 7. Reservoir areas in the Smoky Hill River Basin, in Kansas. The other includes Enders, Harlan County, Medicine Creek, and Wray Reser- voir areas in the Republican River Basin, in Nebraska and Colorado. Also in progress and nearing completion was a report summarizing paleontological survey data on 14 river basins containing a total of 68 reservoir areas. Supplementary appraisals based on further archeological field surveys during the summer of 1947 were being prepared for Boysen, Canyon Ferry, Glendo, Oregon Basin, and Tiber. First drafts of technical reports were under way or completed for Anchor, Boysen, Devils Lake, Enders, Glendo, Jamestown, Kortes, Medicine Creek, Oregon Basin, Sheyenne, Harlan County, and Kanopolis. Since these are based on preliminary reconnaissance which, in some cases at least, will be followed up by intensive survey and excavation, their publica- tion at this time or in the foreseeable future is not contemplated. FIELD WORK AND EXPLORATIONS KANSAS AND COLORADO In northern and northwestern Kansas and northeastern Colorado, four proposed Bureau of Reclamation reservoir projects were sur- veyed for archeological remains by a River Basin Surveys party. The party consisted of Wesley L. Bliss and J. J. Bauxar, who were in the field on this assignment from April 24 to May 7. Projects visited included Wilson Reservoir, in the Smoky Hill Basin in north-central Kansas; Norton and Pioneer Reservoirs, in the Republican Basin in northern Kansas; and Bonny Reservoir, in the same basin in north- eastern Colorado. The investigations were all of preliminary char- acter, and in no case was complete coverage of the proposed pool area possible. Further and more intensive surveys are recommended if and when reservoir construction is undertaken. Wilson Reservoir—tThis area is located on the Saline River, in Russell County, Kans., with the dam site near the eastern edge of the county. Since the future pool area will be some 25 or 30 miles long, the 14 days allotted to reconnaissance obviously permitted spot checking of only a very small portion. The six sites recorded thus represent but a fraction of the total to be expected when intensive survey is made. Of the six sites recorded, three represent pictograph localities. At one, in addition to carvings of Indian origin, were found the names of William F. (Buffalo Bill) Cody and Wild Bill Hickok. The authenticity of these remains to be determined. Two other sites were 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT apparently occupational areas, that is, camp or village locations. At still another, a buried stratum yielded animal bones and charcoal, but no artifacts from which the nature and relationships of the horizon might be suggested. There is a possibility of some antiquity for this material, but further testing is needed. Norton Reservoir.—This is on Prairie Dog Creek, a southerly affluent of the Republican River, in Norton County, Kans. The area to be inundated is small, but has archeological interest. Three sites were located, all characterized by quantities of worked yellow jasper and rejectage. There was no pottery on any of these, or elsewhere in the sections visited. A small depression on one site possibly indi- cates a former pit house. The terrain in the reservoir area generally appears favorable for former Indian utilization. Prairie Dog Creek enters the Republican less than 35 miles to the northeast, within the Harlan County Reser- voir which is now under construction by the Corps of Engineers. About the junction of the two streams are numerous sites represent- ing not less than four pottery-making cultures of varying antiquity. A few miles east of the proposed Norton Reservoir is an aboriginal quarry from which the Indians obtained limestone for use in pipe making. All this leads to the suspicion that further and more in- tensive survey will disclose a number of additional archeological localities at Norton which will be affected by the proposed water-con- trol developments. Pioneer Reservoir.—Located on the Arikaree River in Cheyenne County, Kans., this will affect an area extending southwestward across the State line into Yuma County, Colo. A single site was found here, on the western terminus of the proposed dam axis. Insufficient material was collected from it to make possible a suggestion as to relationships to known archeological complexes of the region. Bonny Reservoir.—This is to be on the South Fork of Republican River, in the southeastern corner of Yuma County, Colo. Though of small extent, it disclosed three sites of archeological interest. Arti- facts of the Yuma horizon, an early prepottery complex, are reported to have been found on one by a local collector, and there appear to be cultural deposits remaining which would be worthy of excavation. Two other sites yielded too little material to be identifiable, even tentatively, as to people or period. No pottery was found within the future reservoir area. NEBRASKA An important share of the 1947 archeological field work of the River Basin Surveys went into reconnaissance and excavation in water- NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 9 control projects in Nebraska. Preliminary reconnaissance included five proposed reservoir areas in the Republican drainage in the south- western part of the State, and six in the Lower Platte Basin in the central portion. The Republican River locations were surveyed by W. L. Bliss and J. J. Bauxar between April 24 and May 7; units visited include Beaver City, Red Willow, Culbertson, Rock Creek, and Buffalo Creek. Results of this work, other phases of which have been noted in the section on Kansas and Colorado, are presented first in the summary which follows. In the Lower Platte Basin, from May 3-12, M. F. Kivett and J. T. Hughes carried on reconnaissance at Amherst, Buffalo Creek, Cairo, Ericson, Mullen, and Rockville units. These projects are all under the Bureau of Reclamation. In addition to the above, archeological excavations were begun at Medicine Creek when it was learned that early construction was planned by the Bureau of Reclamation. Work here was carried on from September 10 until November 9 by the River Basin Surveys, in continuation of previous excavations between July 25 and Sep- tember 10 by the Nebraska State Historical Society, a cooperating agency. In this section only the later researches will be summarized ; operations of the Historical Society are noted further in another place. REPUBLICAN RIVER BASIN Previous reconnaissance by the River Basin Surveys, together with earlier investigations by the Nebraska State Historical Society and the University of Nebraska, have shown that the Republican River watershed contains abundant, varied, and important archeological remains. From 1777 or before until the early 1800’s, the Pawnee lived in one, two, or more earth-lodge villages along the Republican, where it crosses the Nebraska-Kansas State line. At one of these, in the present Webster County, Nebr., Pike visited in 1806. The area to the west contained no permanent settlements and was principally hunting range for the Pawnee, Dakota, Cheyenne, and other tribes. In earlier days, preceding arrival of white men, there were several successive occupations of the valley. Remains attributable to earth- lodge-using, semihorticultural Indians are widely scattered throughout the area, occurring westward nearly to, or perhaps beyond, the Nebraska-Colorado State line. They occur to some extent in the main valley, but are more plentiful on the tributaries. Between this occu- pation, estimated to have taken place during the thirteenth to fifteenth centuries, and that of the historic Pawnee, there were at least two others by peoples who made pottery and perhaps practiced horti- IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VORA ure culture. Still earlier was a fifth pottery-making group, the Wood- land peoples, for whom there is as yet no conclusive proof of corn growing. There is also a growing body of evidence to show that much older, prepottery horizons are probably present, and that such ancient big-game hunters as the Folsom, Yuma, and perhaps other peoples as yet unrecognized or unnamed, passed through the district at various times in the remote past. In short, the east-west valley of the Repub- lican, were its prehistory systematically and thoroughly worked out, would in all probability yield an exceptionally useful and important archeological cross section for the central plains region. Noteworthy, too, is the fact that at several points in the valley prehistoric remains are found in buried soil zones, covered by what appear to be wind-laid deposits. Sometimes two or more such buried zones, lying one above the other, contain markedly dissimilar artifacts attributable to distinct occupations, and are separated by culturally sterile strata of varying thickness. Here is a suggestion that the successive inhabitants perhaps entered the region during climatically favorable times, only to be forced out of it toward the east during periods of deficient rainfall, which are marked by the sterile over- lying dust deposits. Since it is to be expected that the archeological horizons will some day be datable in terms of our calendar, we may hope further to get some concrete indication of the time when these presumed prehistoric droughts transpired. An exceptional oppor- tunity here awaits combined attack by archeologists, geologists, soils experts, paleontologists, and students of other disciplines. Beaver City Reservoir.—The area here involved is on Beaver Creek, in southeastern Furnas County. Beaver Creek enters the Republican from the southwest at a point within the upstream limits of the Harlan County Reservoir, now under construction. Four sites were recorded here. All were littered with quantities of worked and unworked yellow jasper, and were closely similar to others found in Norton Reservoir immediately to the south in Kansas. There were no pottery remains, and it is impossible to suggest the age or cultural affiliations of the sites. They suggest workshops, and may be of no great antiquity. Buffalo Creek Reservoir—tThis is on Buffalo Creek, a northerly tributary of the Republican in Dundy County. The single site located was in a cultivated field, the surface of which was littered with village refuse. Some tendency toward concentration of the remains in smaller areas was noted. Pottery fragments suggest either Upper Republican or Woodland types, and there was a wide variety of stone artifacts. The site seems important enough to warrant further investigation. Culbertson Reservoir——Located on the Republican River in east- NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL II central Hitchcock County, this area yielded only one site in the single day spent here. Yellow jasper artifacts and rejectage, together with the apparent absence of pottery, recall the similar localities found in Furnas County, Nebr., and Norton County, Kans. It is quite probable that additional sites occur within the future pool area. Of historic interest in the area is Massacre Canyon, where the last engagement between the Pawnee and Ogallala Sioux took place on August 5, 1873. The burial ground of the Pawnee slain at the time will probably lie within the reservoir pool. Red Willow Reservoir.—This is to be located on Red Willow Creek, in Frontier and Hayes Counties. Two alternate dam locations have been proposed. Five archeological sites were found, all with pottery remains. Tentatively, the ware is assigned to the Upper Republican horizon. At one place, there was evidence of a habitation structure at the edge of a cut bank. It seems probable that the area is a prolific one archeologically, and that among its prehistoric inhabitants were included several communities of settled, corn-growing peoples. Further work is recommended. Rock Creek Reservoir—tThis locality is in south-central Dundy County, on Rock Creek, a small tributary of the North (Ari- karee) Fork of the Republican. It is quite small, and but a single site was found. From the fact that it is buried beneath some inches of overburden, the possibility of a moderate antiquity is suggested. No artifacts of diagnostic character were obtained, and test excavations will be necessary before any suggestions as to age, relationships, or possible importance can be made. Lower PLATTE BASIN Water-control projects proposed for this subdivision include a num- ber of localities on the Loup River and its tributaries. Most of the area, and the great majority of projects so far announced, lie north of the Platte, in a section of rolling loess hills. The headwaters of the Loup system are in the Sandhills region, but the greater part of the stream valleys flow through a fertile tract of loessial soils. In historic times the area was controlled by Caddoan- and Siouan- speaking tribes who lived in fixed earth-lodge villages; grew corn, beans, and squash; made pottery; and practiced a variety of other arts and industries associated with a reasonably settled mode of life. Since, apparently, their earliest contact with white men, the Pawnee resided in a series of large, fortified towns along the lower Loup, centering in the present Howard, Nance, and Platte Counties, and on the nearby I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. ert reaches of the Platte. To the east, chiefly along or near the Missouri, dwelt the Omaha; upstream, also on the mainstem, were the Ponca; the Otoe and Missouri lived on the Platte below Elkhorn River and elsewhere in southeastern Nebraska. On the west, the Lower Platte Basin was hunting range for the still unidentified Padouca before 1800, and for the Dakota and other roving, nonhorticultural horse nomads after that date. As in the Republican drainage, so here it is evident that before arrival of the tribes named above, a succession of other native peoples had lived in and traveled through the region. From the surveys here summarized, and work previously done at several localities by the Nebraska State Historical Society and the University of Nebraska, at least three earlier occupations by pottery-making and possibly or probably horticultural peoples may be recognized. These include remains attributed to one or more variants of the Woodland horizon, regarded as perhaps the earliest pottery-making peoples of the area; the Upper Republican peoples, who dwelt in small, unfortified villages of earth lodges, raised corn and beans, and otherwise left evidences of a relatively stable tenure of the land; and a still unidentified but apparently later group, the nature of whose occupation remains to be disclosed through archeological excavation. On the headwaters of the Middle and North Loup are village sites attributable to the Dismal River complex, also of fairly late date and probably of Apache or Comanche origin, but not yet identified with certainty with either of these. The principal archeological excavations to date in the area with which we are here immediately concerned are few indeed, considering the extent and abundance of remains. They include work by the Nebraska State Historical Society on Davis Creek, in Greeley, Howard, and Sherman Counties; and on Myra Creek, in Valley County ; by the University of Nebraska in Sherman County; and at various times by both organizations along the Loup River, in Howard, Nance, Platte, Colfax, and other counties along the major river valleys. Reports have been issued on the findings in Valley and Sherman Counties, and on some of the work along the Loup, but a great deal of the material remains unpublished. Amherst Reservoir—This is proposed for the Wood River, a small branch of the Platte, in west-central Buffalo County; the dam is to be at the south edge of the town of Amherst. The terrain is well suited to aboriginal occupancy, and the area is evidently one of considerable archeological interest. Five sites were located and recorded; all are within the proposed pool area. Three are represented by pottery NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 13 remains probably attributable to the Upper Republican period. At one of these there are shallow depressions in uncultivated ground, suggesting pit-house ruins. Two sites yielded thick, coarsely tempered, cord-roughened sherds similar to those of the Valley focus of the Woodland period. Here, as with many Woodland sites, the cultural level seems to be quite thin, but there is a possibility of pits and other underground features. One site, 25BF2, yielded sherds of both Woodland and Upper Republican types, suggesting two occupations and the possibility of stratification of cultures. The Upper Republican remains seem to show some variations from the usual run of this material, but whether the indicated differences are due to time or other factors cannot now be surmised. Local collectors report the presence of many other sites within the reservoir area, and further intensive survey is called for. Buffalo Creek Reservoir—tThe locality here involved lies in north- central Dawson County, approximately 9g miles northwest of Lexing- ton, on Buffalo Creek. Despite a generally favorable terrain, only one site was found. There was no pottery, and assignment to any known cultural complex is at present impossible. It is to be noted that many of the most promising terraces bore a heavy cover of prairie grass at time of the survey, and there may therefore well be other locations as yet unfound. Cairo Reservoir.——This unit is proposed for Dry Creek, in Hall County, approximately 1 mile above Cairo. The area is small, and the topography not especially inviting from the standpoint of abo- riginal occupation. No sites were found, and it appears improbable that further work will be needed, unless construction operations reveal now-unknown materials. Ericson Reservoir—tThis is proposed for Cedar River, in Greeley and Wheeler Counties, near the eastern edge of the Sandhills. Five sites were located, all within the proposed pool area. Pottery of rather distinctive character was present at all. The sherds were small, of a hard, thin ware composed of fine-textured gray paste with a small amount of grit temper. Exterior surfaces are generally plain; others have been treated with a simple carved paddle, and a few are cord- roughened. Parallel trailed lines above a sharp shoulder, and flaring rims with short diagonal incisions on the inner lip, were present. There is one strap handle with zigzag incised decoration. Projectile points are small triangular, with or without lateral notches; small scrapers and beveled-edge knives also occur. The remains are quite similar to others excavated by the River Basin Surveys in 1946 at a village on Prairie Dog Creek, in I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS LEE Harlan County Reservoir, and also to some from sites in Holt, Knox, and other counties in north-central Nebraska. At Harlan County, limited tests revealed shallow, vertical-walled storage pits usually filled with bison bones and other domestic debris. Typologically, the specimens suggest a late prehistoric or very early protohistoric com- plex, and one is tempted to wonder whether it may represent the trail of the Arikara or an early ancestral Skidi-Arikara group moving from northern Kansas through east-central Nebraska to the upper Missouri. If the makers of this material left lineal descendants among the known historic groups of the area, a Caddoan or possibly some Siouan people would seem the most likely possibility. This, obviously, is highly conjectural; but the problem awaiting study here would seem an interesting and important one, and the point of attack readily apparent. Of quite dissimilar nature were a few thick, coarsely tempered, cord-roughened sherds apparently attributable to a Woodland complex. Mullen Reservoir.—This locality is far up the Middle Loup River in northeastern Hooker County, well within the Sandhills region. The proposed dam site lies about 5 or 6 miles east of Mullen, and the reservoir will be approximately 7 miles long. Eight sites of varied age and origin were recorded. Judged by the surface collections made, four sites are attributable to the Dismal River complex, of late pre- historic or early protohistoric age, and one to a variant of the earlier Woodland culture. Another yielded a few sherds of unidentified cul- tural affiliations, similar to those at the majority of sites recorded from Ericson Reservoir. The remaining sites yielded only stone and bone implements, which perhaps represent still another horizon; a cache of more than 140 chipped artifacts was found eroding out of the bank at one of these. At least two of the Dismal River sites are rather extensive, and show certain areas that would undoubtedly repay excavation. There has been virtually no excavation at sites of any horizon in the Sandhills region. Rockville Reservoir.—This is on the Middle Loup in southeastern Sherman County, between Rockville and Loup City, and a few miles above the confluence of the Middle with the South Loup. At time of the reconnaissance, many of the more favorable terraces had been freshly plowed, and conditions were unsuited to site hunting. Nothing of archeological interest was located in the available time. It is be- lieved, however, that further and more intensive search might be worth while if and when construction is initiated. | NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 15 MeEpDICINE CREEK RESERVOIR Medicine Creek Reservoir will be located on Medicine Creek, about 84 miles above Cambridge, in southeastern Frontier County. With public announcement during the summer of 1947 that contracts for this unit would be let within a few months, the River Basin Surveys took steps to salvage certain archeological materials threatened with early destruction. Preliminary surveys had been made here during the 1946 field season by Kivett and Shippee, who recorded 15 sites within the reservoir area. Two promising village sites, marked by potsherds, animal bones, mussel shells, stone artifacts, and rejectage, were situated on the west abutment of the proposed dam; two others, with similar evidences of relatively permanent occupation, lay on the left bank of the creek, at or very near a proposed borrow area just above the dam site. Because only very limited excavations had pre- viously been made in the area, which seems to have been rather thickly settled in prehistoric times by Upper Republican peoples, two members of the Surveys staff were detailed to conduct investigations at and near the dam site. This work began early in September and continued until November 9; it was directed by M. F. Kivett, assisted by George Metcalf, and such local labor as was obtainable from time to time. As previously stated, the River Basin Surveys investigations in 1947 were a continuation of excavations begun on July 25 by the Nebraska State Historical Society. Test excavations were made by the River Basin Surveys party in an occupational area, 25 T18, on the left bank of Lime Creek near its junction with Medicine Creek. A trench 15 feet wide was cut through the site from the south edge northward for 55 feet. Village debris varied in thickness from 12 inches at the south to approximately 30 inches at the north. Hearth areas consisting of burned red earth underlying ash beds were associated with shallow circular pits dug into the sterile yellow subsoil. Unworked fresh-water mussel shells, stone artifacts, and bone fragments were common throughout the fill. Stone artifacts included several small, stemmed projectile points, knives, and scrapers. A few awls and tubular beads of bone, as well as shell disk beads, were found. Pottery, though not plentiful, was of distinctive character; it included thick, cord-roughened body sherds, usually tempered with calcite, and one straight, undecorated rim fragment. The material, in general, is suggestive of the Woodland variant known as the Valley focus, from excavations by the Nebraska State Historical Society in Valley County, Nebr. The pottery shows also some char- acteristics of Woodland materials from Lane County, in west-central 2 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Kansas. The present sample is small, however, and does not permit definitive conclusions. At site 25FT39, situated on a high terrace on the right bank of Medicine Creek about 3 miles above the dam site but within the future’ pool area, two earth-lodge floors were cleared. Both had apparently been burned on abandonment, and contained large quantities of broken pottery. This included several restorable vessels which had been resting on the floor, bottom side up. The interior of one of these had been coated with a red film. Fragments of two human skulls came from a cache pit in house 1, and portions of another were found in a cache in house 2. Grinding slabs for crushing corn or seeds, and made of limestone, were found in both houses. In addition to the pottery, there were stone, bone, and other artifacts. Two similar house floors and a midden area were excavated at site 25FT17, which lies atop a ridge that will be a part of the west end of the dam. House 1 yielded comparatively few artifacts, other than a representative series of potsherds of Upper Republican type. Unusual is the finding of one rim fragment with shell tempering. Two coarsely tempered, cord-roughened body sherds of apparent Woodland type, were found slightly below the house floor level, sug- gesting that remains of an earlier occupation may underlie the ruins of the earth-lodge village. House 2 yielded much broken pottery of Upper Republican types, as well as an abundance of bonework, in- cluding awls, bison-scapula digging tools, a fish hook, and other objects. Of interest were two ground-stone celts, as well as the usual chipped forms. A refuse area approximately 250 feet northeast of house I was dug, and from it was taken a good series of Upper Republican pottery, bone, and stone specimens. Another extensive refuse deposit on the slope east of house 2 was not opened. With the work done by the Nebraska State Historical Society earlier in the season, seven house sites were opened by year’s end in the lower Medicine Creek Reservoir area. All may be attributed to the Upper Republican horizon; variations from site to site may be due to time or group differences, or may represent merely inadequate sampling. Charred kernels of corn occurred at nearly all sites, which is in line with what we know of the semihorticultural practices of the people. Charred post and beam samples, and some badly decayed post sections, were collected, from which it may be possible ultimately to date the occupation. The remains found during the above researches represent only a fraction of those that will be affected by reservoir construction. It is already obvious that not less than two periods of occupancy by pottery- NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 7 making groups are manifested, and there is good evidence that much older materials probably representing prepottery peoples are also present. The proximity of several earth-lodge village sites to the dam- construction area seems to offer an exceptional opportunity for study of one or more prehistoric communities of corn-growing Indians near the presumed western limit of aboriginal American agriculture in the Great Plains. The stripping by power machinery of the topsoil in preparation for laying of the earth dam fill may lay bare most of the former village areas, thus making possible the close scrutiny and precise mapping of most or all of the house units, storage pits, refuse deposits, and other features associated with the former human occu- pants. Such mapping of the community plan, combined with recovery of a large sample of the artifacts and other remains from all sections of the village, would permit a more definitive analysis of a prehistoric settlement than has yet been made in any part of the Great Plains. Such studies at two or three of the sites that will be worked over by power machinery, supplemented by limited tests at other sites in the reservoir area, are urgently recommended as a corollary to the actual dam-construction program. It should be remembered that some of the prehistoric sites in, above, and below the future pool area, like many in other parts of the Repub- lican River watershed, are buried beneath prehistoric dust blankets. The correlation of native occupations, as determinable by archeological methods, with the cyclical or other periods of deposition represented by the intervening or overlying soils, promises to give a sound footing for any attempts at dating prehistoric climatic fluctuations. Such problems, of course, transcend the field of archeology, and call for an interdisciplinary attack by geologists, soils experts, and other specialists, as well as by archeologists. SOUTH DAKOTA During the calendar year 1947, archeological work by the River Basin Surveys in South Dakota was restricted to Fort Randall Reser- voir in the immediate valley of the Missouri River. The field unit was led by Paul Cooper, assisted by J. J. Bauxar, with Robert L. Hall and Warren Wittry as student helpers. The party left Lincoln on June 3 and terminated its activities during the first week in November. The first 6 weeks were devoted to a rapid survey along both banks of the Missouri from Fort Randall to Fort Thompson, to determine in a general way the nature and extent of archeological remains in the area. Beginning on July 18, limited test excavations were made, 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III with such added labor as was locally obtainable, at several of the more promising sites on the left bank of the stream. These included loca- tions in the vicinity of Pease Creek and Wheeler Bridge, and above the mouth of Platte Creek, all in Charles Mix County. The rugged and roadless nature of much of the area, more than 200 miles long, was a serious obstacle, further complicated during the early stages of reconnaissance by a period of excessive rainfall. Fort Randall Reservoir —tThe reservoir under construction here is one of five large multiple-purpose projects planned by the Corps of Engineers for the mainstem in South and North Dakota. The dam site is located in Charles Mix and Gregory Counties, some 5 or 6 miles north of the Nebraska State line, 7 miles south of Lake Andes, and about 60 miles by river above Yankton, S. Dak. An earth-fill structure, it will have a crest length of nearly 2 miles and a maximum height of 160 feet above stream bed. At the planned maximum pool elevation of 1,375 feet (mean sea level), the impounded waters will back up approximately too miles to a point beyond the Big Bend, and will inundate 108,000 acres in Charles Mix, Gregory, Brule, Lyman, and Buffalo Counties. The Missouri River here flows in a flat-floored trench from I to 2 miles wide, bordered by high bluffs which have been moderately to extensively dissected. Alluvial terraces are less common, especially along the right bank, than they are above Chamberlain. White River, entering from the west approximately 11 miles below Chamberlain, is the only major perennial tributary entering the reservoir area. Deciduous timber, chiefly cottonwood and willow, occurs on the main stream flood plain and on tributary valley floors. The bluffs and uplands, where not under cultivation, consist of prairie grasses. Small game is still moderately plentiful, but such once-abundant larger forms as bison, antelope, and elk are now extinct, and deer are nearly so. The Fort Randall Reservoir area is surprisingly little known, archeologically speaking. Prior to the survey work summarized herein, the University of South Dakota Museum provided the River Basin Surveys with a location list of 27 known sites between Fort Randall and Fort Thompson. There appears, however, to be not one extant report of archeological investigations here, although the University — of South Dakota Museum conducted important excavations in 1941 at Scalp Creek and at Ellis Creek, on the west bank of the river. This dearth of well-authenticated field data is the more remarkable in view of the strategic location of the district on the natural line of Indian travel between the Arikara-Mandan habitat on the upper NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL IQ Missouri above Pierre, and the Central Plains area of Nebraska and Kansas to the south. Migration legends link with the Fort Randall district not only the early Arikara and Mandan, prior to their move- ments upstream, but also the Iowa, Omaha, and Ponca, who later migrated downstream, and the Teton and Yankton Dakota on their movements from east to west across the Missouri. In historic times— that is, during the latter decades of the eighteenth and early half of the nineteenth centuries—there were no permanent Indian towns in the area, then dominated by the Teton on the west and the Yankton on the east. The present reconnaissance, by bringing to light a large number of sites previously unrecorded, demonstrates that it is only lack of sustained search and not actual lack of aboriginal remains that has made the section largely a blank on archeological maps of the Great Plains. The River Basin Surveys party located and recorded 93 sites ; and since considerable sections of the river banks were not accessible in the available time, it is highly probable that a number of additional localities of archeological significance remain to be located and in- ventoried. Among those now known are both fortified and unforti- fied earth-lodge villages, stratified and unstratified occupational areas with dwellings of unknown character, mounds, burial grounds, tipi- ring sites, and other antiquities of undetermined nature. In time, they range from those probably or certainly attributable to the recent Yankton Dakota to others, far more numerous, of the prehistoric period—a time span of perhaps 10 centuries or more. In some, there are multiple occupation levels separated by culturally sterile strata possibly indicative of climatic fluctuations, such as droughts. Arti- facts, though often scanty, indicate relationships on several time levels with other peoples and cultures to the north, east, and south. : As with reconnaissance work generally, where little more than surface survey and collecting are done, so here at Fort Randall the artifacts recovered in the 1947 operations are quite limited in quantity. Furthermore, there is no established framework of human prehistory in the district, based on careful analysis of data gathered through con- trolled excavation and laboratory work, into which the new findings may be fitted. Thus, assignment of remains and interpretations as to culture history are not yet possible. Certain clues may be found in the presently available data, however, and it may be worth while, therefore, to note their nature briefly. Particularly noteworthy are the results of test excavations in two mounds (39CH4) located a short distance below Wheeler Bridge on the left bank of the Missouri. This is near the western limit of 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III occurrence of burial mounds, found in increasing numbers farther east in the James and other lesser stream valleys, but exceedingly rare along the Missouri itself. The mounds, greatly reduced by long cultivation, were some 40 to 50 feet in diameter and less than 4 feet high, with circular outline. The larger contained evidences of a pre- pared floor of bluish clay. In each was found a subrectangular pit approximately 5 or 6 feet across, dug a foot or more into the under- lying ground surface. Each pit contained the disarticulated bones of several individuals, presumably interred after exposure of the corpse had destroyed the softer tissues. In the larger mound, besides the suggestion of a prepared clay floor, there was evidence (pl. 1, fig. 1) of a log layer over the burial pit, and some of the leg and arm bones had been perforated near one end—features heretofore unreported from the supper Missouri and Great Plains region. There were no artifacts in association with the skeletal remains. Location of the vil- lage, if any, whose inhabitants built and used the burial mounds, is unknown. Two miles downstream, on a small terrace at the junction of Pease Creek with the Missouri, exploratory pits and trenches disclosed remains (39CH5) attributable to two periods of occupation. In both levels pottery was present, though in small amounts only. That in the lower and earlier deposits appears to be in the Arikara tradition. Potsherds from the upper and later strata, on the other hand, differ, but their relationships are still unclear. A trash mound approximately 45 feet deep includes remains of both periods, thus affording an opportunity for significant and definitive stratigraphic studies. No house remains were detected, though their presence is suspected. Burials are reported to have been uncovered by road-building work on a bluff across the creek to the northwest. Testing operations at the Oldham site (39CH7), about 1 mile above the mouth of Platte Creek, also revealed evidence of two suc- cessive occupations. The earlier, indicated in the diggings as a buried dark-gray soil stratum containing charcoal, flint chips, and animal bone, was represented by fragments of pottery with the surfaces roughened by a cord-wrapped paddle. In the upper zone, the pottery fragments either had plain surfaces or else had been treated with a grooved or thong-wrapped paddle so as to give a ridged effect. Asso- ciated with the second and later occupation were semisubterrean dwellings, two of which were cleared (pl. 2, fig. 1). Each had numer- ous closely spaced post molds outlining a well-defined circular floor; a narrow, formerly covered entrance passage; a central fireplace; and four primary posts supporting the main house structure. Many NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 21 of the molds yielded quantities of reddish-brown decayed wood and wood dust, some of it identified through its characteristic odor as juni- per. Basically, the house type indicated is not unlike that used by the Pawnee, Arikara, and other early historic corn-growing, earth-lodge- dwelling Indians of the eastern Great Plains, though it deviates in cer- tain particulars. Several refuse and cache pits were also opened. A ditch for defense still partially encloses the village area of perhaps 4 or 5 acres; cultivation has presumably obliterated this feature over the rest of the site. The ditch was probably built by the people who lived in the circular dwellings, and so belongs to the later occupation. One is tempted to suggest that the second and more recent occupation of the site may be attributable to the late prehistoric or early historic Arikara on their way up the river, although there is always a possi- bility that other tribes may have shared this particular grouping of material culture traits at one period. Further and more extensive excavation is called for. On none of the sites tested was any evidence found of contact between Indians and white men. All, as indeed the great majority of those recorded, will be flooded or otherwise adversely affected by the reservoir and associated works. NORTH DAKOTA Two major water-control projects in North Dakota were surveyed for archeological remains by a River Basin Surveys field party. Marvin F. Kivett was in charge of the field work; he was assisted by Gordon F. McKenzie, John L. Essex, and Leo L. Stewart, students. From June 13 to August 19, this group made a preliminary recon- naissance in the Garrison Reservoir area, on the Missouri River above Bismarck. On August 21, activities were transferred to Baldhill Reservoir on the Sheyenne River above Valley City, where work was terminated on August 29. At both projects, construction is under way by the Corps of Engineers, and the time available for locating, record- ing, and salvage of archeological materials is rapidly becoming shorter. Garrison Reservoir—This project, one of the largest proposed or under construction for the Missouri River Basin, is in the north- western part of the State. The dam site is on the mainstem in McLean and Mercer Counties, 55 miles northwest of Bismarck and about 15 miles south of Garrison. Here a rolled-fill earth embankment more than 2 miles long will rise 210 feet above stream bed to create an artificial lake extending upstream to a point above Williston, nearly 200 river miles distant. At the planned maximum pool elevation of 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III 1,850 feet (m.s.l.), some 390,000 acres will be under water. Affected directly will be portions of Mercer, McLean, Mountrail, Williams, McKenzie, and Dunn Counties, and of the Fort Berthold Indian Reservation—an area equivalent to nearly half that of the State of Rhode Island. Comparatively little is on record regarding the prehistory of this section of the Missouri. The reservoir will lie for the most part upstream from the area generally identified in historic times with such Upper Missouri village tribes as the Mandan, Hidatsa, and Arikara. In sharp contrast to numerous sites along the mainstem from Knife River to the White, where even the casual visitor may see house depressions, cache pits, fortifications, refuse mounds, and other sur- face traces, the remains above Garrison dam site are usually small, more or less deeply buried, and quite inconspicuous. For the Garrison Reservoir area, approximately a dozen sites were reported to the River Basin Surveys field party before it began operations. Most of these were thought to be winter villages, temporary camps, or late sites showing few of the pre-white elements of Indian culture. Coverage by Kivett’s party included areas in five counties, above and below Fort Berthold Indian Reservation. No work was possible on the Reservation, but record was made of nine sites reported to exist thereon. Areas most readily accessible by automobile received the closest attention. In all sections of the area seen, however, it is believed that further work is needed; and it is anticipated that more intensive survey, including access to the Reservation lands, will add many other sites to the present list. Flint chips, stone, bone slivers, and other evidence of former human activity are to be found on virtually every suitable terrace throughout the area. In the 2-month survey, 70 sites and localities were recorded, most of them apparently unknown previously. In- cluded are 59 occupational areas, I burial site, and 10 unclassified locations. No burial mounds were noted, but local informants report occasional burials in rock piles on some of the tipi-ring sites. Eleven of the occupational areas consist of grouped circles of glacial boulders, to to 20 feet in diameter, and located usually on the bluffs and uplands above the future water level. Locally these are termed tipi rings, on the supposition that the stones were used to hold down skin tipi covers. Such sites are particularly common on the left bank of the future reservoir area between Sanish and Williston. Refuse and artifacts are usually scarce about these sites ; when present, they include flint chips, occasional arrowpoints. scrapers or knives, grooved mauls, and perhaps glass beads and metal. Presence of metal NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 23 and glass suggests recency of occupation, but there is so little asso- ciated cultural material that no assignment of the sites to a known historic or other group or groups is now possible. Indicative of a more settled mode of life are several sites similar to the ruined villages found in such impressive numbers farther down- river. These are not plentiful, but they occur both below and above the Fort Berthold Indian Reservation ; others are reported to exist, as might be expected, within the Reservation boundaries. Circular de- pressions mark the sites of former earth-covered lodges, or of under- ground storage pits. Refuse is more plentiful on these sites, and includes broken pottery, worked stone, animal bones, etc. At least one such site, the Rock Village (32ME15) on the right bank just above the abandoned town of Expansion, appears to have been sur- rounded by a protective ditch. One of the circular depressions here was tested, disclosing a slab-lined fireplace 10 inches below the sur- face. Materials seem to be rather plentiful on the surface; those recovered include only articles of native workmanship—pottery, worked stone, steatite vessel fragments, animal bones, and a grooved maul, ‘Pottery appears to be in the Mandan-Hidatsa tradition. North Dakota workers have suggested a pre-1850 Hidatsa origin for the site (pl. 4, fig. 1). _ Of more recent date is another earth-lodge village (32MZ1), oppo- site the mouth of Little Knife River. Known as the Crow Flies High village, it is believed to have been occupied between 1868 and 1893 by the Hidatsa. Metal, glass, and other recent materials were plenti- ful, but there was little of native origin. Most of the sites located and recorded by the Kivett party consist of artifact- and refuse-bearing strata covered by a few inches to several feet of wind-blown soil (pl. 3, fig. 1). Such locations, of course, are not usually apparent on the ground surface, and must be searched for along cut banks and eroded areas bordering the stream courses. They are marked by outcrops of burnt earth, ashes, charcoal, occasional hearths, flint chips, animal bone, and sometimes stone or bone artifacts. At least one site in McLean County (32ML9) was found to be stratified. On the surface were a few small smooth and simple-stamped pottery fragments; at a depth of 6 to 12 inches was a mixture of burnt earth, chips, bone, and thick, coarsely tempered potsherds with deep, broad cord impressions. Again, in a site in Mountrail County (32MNg), smooth and simple-stamped sherds occurred from the surface to a depth of 8 inches; a single projectile point was small, triangular, and side-notched. At 12 inches was found a second culture-bearing zone approximately 4 inches thick. This 3 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III yielded thick, coarsely tempered, cord-roughened sherds, together with fragments of large notched and stemmed projectile points. In this and the preceding site, the more deeply buried materials suggest a Woodland horizon, not unlike materials found under somewhat similar stratigraphic conditions in the central plains of Nebraska and Kansas. It is not to be expected, of course, that a relatively rapid surface reconnaissance along several hundred miles of stream bank will permit definite conclusions as to relationships and significance of the materials inventoried and recorded. Nevertheless, it is clear that the Garrison Reservoir area has been inhabited by prehistoric peoples over a con- siderable period of time. The buried sites suggest small groups, some without pottery, others with pottery and perhaps some knowledge of horticulture. It is worth noting in this connection that at the time of first white contact, according to Will, the northern limit of abo- riginal corn growing in the Missouri Valley was probably the Knife River. Kivett observes that village sites appear to have become more permanent and larger in later times, suggesting better adaptation to the rather harsh environment. To what extent the adoption or im- provement of corn agriculture may have figured in this improved living is still uncertain. The soil which covers many of the sites suggests extensive wind action, perhaps correlated with decreased rainfall or prolonged drought conditions. As to succession of occupations, it seems probable that the later sites with earth-lodge circles were left by the Mandan, Hidatsa, or Arikara, but of what period is not always certain. Some are probably late and decadent ; others may possibly represent westward extensions of the vigorous village community economy flourishing farther downstream during the eighteenth century and perhaps earlier. A few pottery sites, to judge from the sherds, are of Woodland origin ; their occurrence in buried zones and stratigraphically below sherd areas of apparent Mandan-Hidatsa affiliation is an interesting parallel to successions already known farther south. Some of these Woodland materials occur on buttes (pl. 3, fig. 2) and other elevated locations. Still other sites, without pottery and under several feet of over- burden, suggest one or more prepottery occupations. No evidence of geologically ancient remains, that is, of Early Man, has yet been recognized in the Garrison area. Baldhill Reservoir.—This project is under construction in east-cen- tral North Dakota, just outside the Missouri River watershed. An earth-fill dam 57 feet high by some 2,000 feet long will be erected on Sheyenne River about 11 miles above Valley City. The reservoir pool, NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 25 approximately 20 miles long, will inundate land in Barnes and Griggs Counties. Normal summer pool elevation is planned for 1,266 feet. In the week allotted to reconnaissance at Baldhill, only a small portion of the future reservoir area was examined. Heavy vegetation doubtless obscured many traces of native activity and concealed sur- face evidence. Nevertheless, it is abundantly clear that the area has been occupied, probably at more than one period, by aboriginal groups who left several types of remains. Ten sites, all previously unreported, were located and recorded by Kivett’s party. Six sites are occupational areas, that is, are presumed to mark areas of former domestic village activity. They occur on low terraces along abandoned stream channels, and are commonly covered by a black alluvial mantle. There are few cut banks or erosion scars where deeply buried strata can be sought. Tests on the terraces showed that some of the occupational areas have considerable pottery, bone fragments, stonework, and other cultural debris. A variety of surface treatments and other techniques is shown by the pottery, and this, with certain other lines of evidence, would seem to indicate that more than one group was in the region. In addition to the camp or village sites, three mound groups were located. The groups consist of two to five mounds each, circular or elliptical in ground plan, 3 to 6 feet high, and 15 to 30 feet in diameter. All are on the bluffs above the proposed pool level; they will not be flooded, but are subject to vandalism by visitors, workmen, and others. Some have already been partially destroyed in this fashion. It is reported that numerous burials have been taken from some of these mounds. The River Basin Surveys party made a small exca- vation in one (32GGr1) in Griggs County, where cultivation had brought human skeletal remains to the surface. The disarticulated skeletons of eight individuals were recovered. Their distribution suggests a cumulative process of mound growth, with burials prob- ably added from time to time, rather than a single mass grave. There were no associated artifacts to indicate possible cultural connections. From the limited samples of artifacts collected, most of them per- force on the surfaces of sites where the chances are good for admixture with later materials, a succession of occupations seems indicated. Some of the pottery fragments show plain unmodified surfaces; others have been roughened through application to the wet clay of a cord-wrapped paddle; still others have parallel ridges, made by treat- ment with a grooved paddle. There is also considerable variety in the tempering material used. The fact that different methods of surface treatment and a variety of tempering materials sometimes 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III occur on a single site under conditions suggesting their use by a single group, may mean that there was in this area a relatively late fusion of cultural practices in vogue among several peoples at an earlier date. Baldhill Reservoir area, and in fact the Sheyenne Valley in general, is one of considerable archeological promise. During the historic period, Indian groups seem to have traveled from east to west across the region. A number of historically important northern plains tribes are thought to have entered the Great Plains from central Minnesota and eastern North Dakota. One such group is the Cheyenne, his- torically a hunter tribe, but as recently as 1770 a settled semihorti- cultural pottery-making people living in earth-lodge villages on the Sheyenne River. A fortified town site attributed to this tribe was excavated in 1938 near Lisbon, N. Dak., by a Columbia University- North Dakota Historical Society expedition. Historical documents or tribal traditions link with the Minnesota woodlands such other tribes as the Siouan-speaking Hidatsa, Teton Dakota, and Assiniboin. Their archeological antecedents are unknown; none have been cer- tainly correlated with any of the several known archeological com- plexes so far recognized in Minnesota and adjacent regions. The valleys of the Sheyenne, Big Sioux, and James Rivers, lying athwart any route westward from Minnesota, should show traces of the passage of migrating groups. One would logically expect that a geographically intermediate region such as the Baldhill locale, a con- venient stopping point for tribes on the move, might show some of the cultural readjustments made in the change from an eastern wood- land to a western plains habitat. The historic tribal movements of the area, and their archeological implications, are but one of several problems to be expected at Bald- hill. Another is the matter of the earth mounds with which the region abounds. Such remains occur in great numbers in northern Iowa and Minnesota. Westward, they are found into the Dakotas in diminishing numbers, reaching the Missouri sparingly in southern South Dakota and elsewhere, but not in general occurring beyond the Coteau du Missouri. So far as known, most of these appear to be burial mounds, often with grave goods. Little in the way of field research has been contributed during the last 40 years toward the matter of age, origin, and meaning of the mounds, and their connec- tion, if any, with the village sites of the region. It seems improbable that the mounds are all assignable to a single people or period; they may well have been constructed over a considerable interval of time, although it is true that none of those so far explored have given evi- NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 27 dence of any great antiquity. The construction of such works, often in groups of three or more, would not have been undertaken by roving bands of hunters. They suggest, rather, fairly permanent village groups. The exact relationships between the mound-building Indians of the eastern Dakotas and the historic Upper Missouri village tribes remain to be worked out. WYOMING AND MONTANA Approximately one-third of the Missouri River watershed lies within the borders of the present States of Wyoming and Montana. The greater portion of this section consists of short-grass plains, a continuation of the grasslands of the western Dakotas and Nebraska. On the west, about the headwaters of the Missouri and its tributaries, the grasslands abut on the Rocky Mountains with their pine forests. In central Wyoming, where the northwest-to-southeast mountain barrier is interrupted, the short-grass gives way to sagebrush plains. Throughout all this great region, deciduous trees occur almost wholly as thin straggling belts along the stream valleys, with coniferous forests on the mountain masses. In historic times, the plains and valleys were occupied by bison- hunting, horse-using nomads. North of the Missouri were the Black- foot, Gros Ventres, and Assiniboin. On the Yellowstone were the Crow ; west and south of them were the Shoshoni. In eastern Wyo- ming were the Teton and Ogallala Dakota, and farther south, the Cheyenne and Arapaho. For most of these, there is good evidence of relatively recent arrival in the region. Concerning the movements of the western tribes, such as the Shoshoni, little information is avail- able for the period before about 1800; and their wanderings before the eighteenth century are largely a matter of inference and conjecture. Incomplete archeological reconnaissance has shown that camp sites, hearth areas, tipi rings, boulder alignments, quarries, and workshop remains are widely scattered over the area. Where geologic conditions are favorable, caves, rock shelters, and pictographs occur. The sites generally are littered in varying degree with chipped-stone work and rejectage; ground stone is not common, though steatite vessels and grinding implements may be found; pottery is rare, and generally absent in the more westerly sites. An interesting, abundant, and as yet little-investigated type of remains is the bison kill, found par- ticularly north of Wyoming and into southern Canada. More recently, with the extensive soil erosion of the recent droughts, the remains of ancient bison hunters have been coming to light in various portions of the region. 28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Systematic investigations have been made so far in only a few spots throughout this vast region. They have shown, nevertheless, that the archeology varies through time as well as through space; that not all the sites can be ascribed to a single period ; that sites exist, both in the open and in caves, where successive occupations are represented by artifact assemblages that show significant changes from level to level ; that while many of the occupations are marked only by relatively thin deposits, some occur under conditions suggestive of a very respectable antiquity; and that influences from several directions have been operative at various times in the past. With a prehistory which apparently extends backward to the time when mammals now extinct roamed the area, the problem of determining the relationships of the pre-horse bison hunters to those of post-Columbian times promises to be intriguing and far from simple. The activities of the River Basin Surveys in the Wyoming-Montana region during 1947 were largely confined to further survey at five Bureau of Reclamation water-control projects. The party doing this work consisted of Wesley L. Bliss, Jack T. Hughes, J. M. Shippee, and H. G. Pierce. Departing from Lincoln on June 10, it operated on the following schedule: at Glendo Reservoir, Wyo., June II to July 2; at Boysen Reservoir, Wyo., July 3 to July 25; at Oregon Basin, Wyo., July 27 to August 11; at Canyon Ferry, Mont., August 13 to August 24; at Tiber Reservoir, Mont., August 25 to September g; and again at Boysen from September 11 to November 6. For these units, preliminary reconnaissance has now been virtually completed, and a limited amount of testing has been done. A total of 236 sites has been located and recorded. In addition, one small cave site at Boysen has been excavated to forestall its despoliation by relic collectors. Boysen Reservoir——Boysen dam, now under construction, is on Big Horn River at the head of Wind River Canyon, in Fremont County, Wyo., approximately 20 miles south of Thermopolis. The crest of the structure will be 140 feet above stream bed ; at normal pool, elevation 4,725 feet (m.s.l.), the reservoir will extend some 20 miles southward along the Big Horn River into Shoshoni Basin to cover an area of nearly 20,000 acres. Excepting the river valley itself, the region is mainly an arid sagebrush desert, with low rainfall, little sur- face water, and few springs. Timber is restricted to stands of cotton- wood along the stream banks, and coniferous trees, mainly yellow pine, on the slopes of the Owl Creek Mountains. In and near the future reservoir area, 75 archeological sites have been visited and recorded by the River Basin Surveys field parties. The sites are varied in character, and, so far as may be judged from NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 29 preliminary examination, they are almost all single-occupation sites ; the one certain exception, a stratified cave deposit, was excavated. Forty-nine sites may be described as camps and occupational areas. They vary somewhat in character of the remains present, in manner of their occurrence, and in topographic location. In general, they are marked by compact to scattered clusters of fire-blackened stones, sometimes occurring in shallow pits and intermixed with bits of char- coal, and surrounded by stone chips, flakes, spalls, occasional grinding stones, chipped artifacts, bone fragments, and similar debris of former human activity. There is no indication of structures, such as tipi rings, about the hearths, which suggests the use of brush or other highly perishable shelters. On some sites debris is relatively abundant; others yield almost none of it. Many of these sites occur in hollows among the sand dunes beside intermittent streams. (pl. ayehig: 2). An interesting group of sites includes thin detrital areas situated in the lee of eroded sandstone outcrops on terraces and low hills over- looking stream valleys. Here there are no stone-hearth clusters or traces of structures; burned stones are scarce; and the remains consist of chipped and broken stone, occasional projectile points, knives, scrapers, or other tools, and weathered bone fragments. Whether these remains are to be attributed to peoples other than those who dwelt in the dune areas, or alternatively indicate some sort of seasonal shift of residence, is not now apparent. Presence of an occasional glass bead or iron fragment suggests that these sandstone outcrops were used as shelters by tribes of the historic period, as probably by others long before. Other remains may be briefly noted. Six tipi-ring sites have been found; at some, stone clusters in the center of the rings indicate the former fireplaces. Stonework and other aboriginal debris is uniformly scarce. Native quarries and workshop sites to the number of seven were located in and near the reservoir area, where quartzite cobbles or other stone materials suitable for tool making were readily avail- able. At six locations there are human (pl. 5, fig. 1) and animal representations pecked into the face of sandstone outcrops; their age and the tribal identity of their makers are unknown. Two rock- - covered burials were found in sinkholes in sedimentary rock outcrops, both with the skull missing. There were no accompanying artifacts, and the tribal or cultural identity of the deceased remains uncertain. Particularly noteworthy among the numerous Indian sites at Boysen is Birdshead Cave (site 48F R54), situated near the base of the Owl Creek Mountains. Within this cave (pl. 5, fig. 2), tests disclosed the 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III presence of several levels of aboriginal occupation. Because the deposit was of small extent and limited depth, and lay close to future developmental lines where looting would inevitably take place, com- plete excavation was undertaken. Excavations confirmed the first impressions of a succession of occupations, and the artifact sample, though small, suggests significant variations from level to level. If the artifacts from the successive strata can be successfully correlated with artifacts from some of the single-occupation sites in the reservoir area, it may become possible to arrange the latter in a sequential order and thus bring a measure of relative chronology into the prehistory of the locality. Analysis of the cave materials is now under way, and the actual significance of the site awaits a full and definitive statement of the laboratory findings as well as the field data. It may be noted, how- ever, that the cave fill consisted of a series of culture-bearing strata separated by layers of decomposed rock and dust. Ash beds and hearths, some of the latter apparently containing fire-blackened stone, were associated with the upper cultural levels. A few plainware potsherds were found on and just below the surface. Below these, but in the upper levels, were steatite vessel fragments, pieces of rabbit-hair cloth, fiber cordage, basketry, small side- and base-notched points, and a few bits of obsidian. At successively lower levels were found small triangular points, then large side-notched points, and finally “fish-tailed” points with basal notch or concavity. Charcoal occurred in abundance at all levels. Refuse animal bone was moder- ately plentiful, consisting mainly of rodent and artiodactyl material, with little or no bison. Strongly suggested in all this is a late prehistoric occupancy of the cave by Indians from the Great Basin to the west rather than by Plains peoples from the east. An interesting set of problems is thus opened up as to the long-time interrelationships between two rather distinctive modes of life in the semiarid western plains. The identity of the poorly represented earlier occupants at Birdshead Cave must remain uncertain until more work has been done at other better- yielding sites. The charcoal from Birdshead Cave is now being studied for dendro- chronological possibilities. Through the courtesy of Gila Pueblo, Globe, Ariz., which placed a Swedish increment borer at the disposal of the River Basin Surveys party, borings were collected from living pines on the slopes about the cave. It is yet too early to indicate the results of these examinations and the likelihood of establishing a tree-ring chronology for the locality. NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 31 Glendo Reservoir —This unit, where preconstruction work is near- ing completion, is located in northern Platte County, Wyo., approxi- mately 45 miles west of the Nebraska boundary. The dam, go feet high, is to be on the North Platte River, about 7 miles southeast of the town of Glendo; the reservoir will have a length of some 12 miles and a maximum surface area of approximately 3,750 acres (elevation 4,590 feet, m.s.l.). The terrain may be described briefly as a rolling to hilly grass-covered plain, with some deciduous trees on the valley bottoms and a few small conifers on the hills and bluffs along the stream. The extensive and important aboriginal quartzite quarries known as the Spanish Diggings lie a few miles to the northeast of the proposed reservoir area. Forty-three localities of archeological interest have been recorded in two seasons of survey at Glendo. Most of these are on terraces or low. bluffs near the river or its tributaries. About half are camp sites, characterized by clusters of fire-blackened stones and areas of occupational debris, such as chips, flakes, and occasional stone artifacts. Some are on the present surface; others lie buried beneath varying depths of wind-blown soils, and have been exposed by gullying or stream action. In general, they show little depth of refuse and suggest short periods of occupancy. Six stratified sites are known. The most promising lie outside the future pool area, but close enough to be subject to extensive vandalism once reservoir construction gets under way. At one, cultural remains have been found to a depth of approximately 9 feet; they include pottery fragments from near the surface and a Folsom-like unfluted point from one of the buried strata. Another such site within the future pool area, 48PL13, showed three successive levels of occupa- tion, at depths of 14, 30, and 60 inches below the surface. The limited artifact collections from these stratified sites strongly suggest cultural variation from level to level, but the quantity of material is insufficient to permit definition of the several complexes apparently represented. The importance of such locations lies in the clues they may give to the sequential arrangement of the numerous other sites in the area where only single occupations are indicated. Six groups of boulder circles representing tipi rings were also located ; they lie chiefly on hilltops, bluffs, and mesas back from the streams. Very little detritus occurs in association, and their age and relationships remain problematical. No caves or rock shelters are known in the reservoir area, though such sites occur in the vicinity. Two quarries were located, both on the hills above pool level. There appear to be no pictograph or petroglyph groups in the area. At one 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III site are a half dozen rock cairns, each approximately 1 foot high by 3 feet in diameter. Their significance is as yet unknown; similar structures have been reported in other localities in association with tipi-ring sites. Although none of the localities now known for the Glendo Reser- voir area indicates any great length or permanence of particular occupations, the evidence strongly suggests repeated use of the area by various peoples throughout a long period of time. Apparently, this use began in prepottery days and continued into quite recent times. Glass beads collected on the surface at one or two sites indi- cate habitation into the historic period, when such tribes as the Teton Dakota, Cheyenne, and others visited the district. Oregon Basin Reservoir—Oregon Basin is a subcircular natural depression about 44 miles in diameter, situated 8 miles southeast of Cody, in Park County, Wyo. Sandstone cliffs and ridges border it on the west and north, with sloping hills on the south and east. The surrounding terrain consists of rolling prairies cut by stream valleys, and barren hills. The flat central part of the basin floor, formerly a lake, will be used for storage of water brought through a 20-mile conduit from Shoshoni Reservoir. The basin floor is at an elevation of 5,100 feet; normal pool elevation will be 5,175 feet, with a surface area of about 4,000 acres. Twenty-eight archeological sites are now on record for Oregon Basin, in and near the proposed reservoir area. They include open camp sites, rock shelters, workshops, and petroglyphs. Some are situated on the beach line of the ancient lake, others on knolls and slopes nearby, still others among rocky outcrops and along creek channels outside the basin. There is considerable variation in artifact types, and the aboriginal occupation of the Oregon Basin region undoubtedly goes far back into the past. The camp sites, 12 in number, consist of scattered clusters of hearth stones, on or near the ground surface, about which is commonly a litter of flint chips, spalls, cores, animal bone, and sometimes projectile points, chipped knives, scrapers, mullers, etc. The quantity of such debris varies greatly from site to site. Pottery appears to be uniformly absent. That these sites are not all from a single period is suggested by variation in form of projectile points. Occasionally these camp sites occur as dark soil zones in cut banks, with burnt rocks, charcoal, and other refuse intermingled. One is situated near a spring on the west side of the basin, and is overlaid by 4 to 6 feet of alluvium; others appear to be weathering out from under sand dunes. In the sandstone cliffs and overhangs which partially surround the NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 33 basin are at least six shelters (pl. 6, fig. 1) formerly used by Indians. In one, the fill containing traces of human activity is more than 10 feet deep; it consists partly of occupational debris, partly of wind- blown sand, and partly of material weathered from the overhanging walls. Though not directly imperiled by the reservoir pool, proximity of these shelters to the future reservoir makes their destruction by relic hunters almost inevitable. Quarry workshops consist of areas littered with cores, spalls, flakes, and blanks ; one was found to the north, another to the south, of the proposed reservoir area. Pictographs occur on cliffs outside the basin. Canyon Ferry Reservoir—Canyon Ferry Reservoir, a multiple- purpose project, will be located on the Missouri River in west-central Montana. The proposed dam site is in Lewis and Clark County, approximately 15 miles east of Helena. From this point, the dam, a concrete structure 175 feet high, will create a lake 24 miles long, extending southward into Broadwater County and covering an area of 35,000 acres at normal pool (elevation 3,800 feet, m.s.l.). The reser- voir site is a broad, fertile valley lying between the Big Belt Mountains to the east and the Elkhorn Mountains to the west. There are numerous sloughs marking former river channels. A wide sloping tableland between river and mountains has been cut at its river front into high steep bluffs. Deciduous trees occur on the valley bottom, with conifers dominating the mountain slopes; otherwise, where not under cultivation, the land is in grass. Thirty-three sites have been located by the River Basin Surveys in and near the future reservoir area. They are generally situated on terraces near the streams, or on the bluffs. Seventeen are camp sites or occupational areas, marked by quantities of chips, flakes, fire- blackened rocks, stone hearths, and occasional stone artifacts. Some lie on the present ground surface ; others occur as detritus-laden strata buried by a few inches to as much as 3 feet of alluvial or wind-de- posited soils. None suggest anything other than a short period of occupancy. The stone-hearth camp sites tend to occur on the flats. Tipi rings also occur, sometimes in association with stone hearths. The rings here are rather small, seldom more than 10 to 15 feet in diameter. They may be closely grouped or widely scattered, but seem to be rather more plentiful on the higher bluffs than on the terraces or valley floors. Many of the rings have been partly destroyed or are almost buried by sod and wind-blown sand. Stone hearths occur sometimes within, sometimes outside, the rings. Artifacts and cultural detritus are very scarce in the vicinity of the tipi rings. No caves, quarries, or workshops were located in the reservoir area. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Poorly preserved paintings in red ocher were found on rock walls outside the future pool area, in localities that will doubtless lead to destruction by vandalism. The limited sample of artifact material recovered consists mainly of stonework. Quartzite predominates, but there are chips of chal- cedony, jasper, obsidian, etc. Projectile points, knives, scrapers, and other implements are found occasionally on the surface. No pottery was noted at any of the locations. Outside the pool area are several sites of considerable promise. At least one of these is said to have yielded points and blades sug- gestive of certain types of early stone industry. It is not unlikely that the occupation of the region here has been intermittent since the days of the paleo-Indian. Further investigation will be necessary before the variant types of points and other artifacts collected sparingly on the shallow camp sites and elsewhere can be arranged in a temporal sequence. Tiber Reservoir—The proposed Tiber Reservoir, for irrigation purposes, is on the Marias River in Toole and Liberty Counties, Mont. The dam site is in Liberty County, 12 miles south of Tiber, and approximately 45 miles above the confluence of the Marias with the Missouri. The dam is planned for a height of 185 feet; it will create a reservoir some 26 miles long, with a surface area of 17,000 acres at normal pool (elevation 2,992 feet, m.s.l.). The terrain is a flat, grassy plateau, cut by small gullies and dropping abruptly into the valley of the Marias. Steep bluffs line the stream at many points. Cottonwood and willow are found on the bottoms along the stream banks; sage- brush covers some of the flats and terraces; grass is characteristic of most of the area. Fifty-three archeological sites have been recorded in and about the Tiber Reservoir area. Most of them fall in one or another of three major categories: buried sites in the river terraces, surface sites on the river terraces, and tipi rings generally located on the bluffs over- looking the river valley. The buried sites obviously precede those on the surface of the terraces; the latter may precede the tipi rings, but of this there is still no definite proof. The buried sites are exposed in cut banks where lateral erosion by the Marias is removing old river terraces. Hearths, some of them apparently consisting of shallow pits filled with fire-cracked stones, ash, charcoal, and blackened earth, and associated with refuse animal bone, flint chips, flakes, and scattered bits of charcoal are to be found at depths of 1 to 23 feet below the terrace surfaces. Some of the over- lying fill suggests river deposition, presumably by the stream in flood ; NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 35 elsewhere, alluvial deposits derived from coulees heading in the bordering bluffs seem indicated. Unfortunately, there is at present no way to judge the rate of this deposition. That no great length of time is necessarily involved is suggested by the finding of blue glass beads and trade iron associated with an ash-filled hearth at site 24TL17, about 10 miles southeast of Shelby, on a habitation level buried beneath 12 to 18 inches of river silts on which stands a scatter- ing grove of large cottonwoods. Despite the rather impressive depth of overburden, it is possible that the rate of deposition in many cases may have been high. Few artifacts were recovered in the survey op- erations at these sites. That many, perhaps most, are prehistoric, is probable; that any now known can be regarded as paleo-Indian is very doubtful. Surface sites on the river terraces consist characteristically of clusters of fire-cracked stones, generally not associated with tipi rings, but with small quantities of worked stone and rejectage scattered in the general vicinity. Many of the hearths are partially buried, and one is tempted to wonder whether, given the proper combination of topographic and climatic factors, these sites would not resemble the buried ones in most particulars. Here the artifact inventory from the two kinds of sites may well be definitive; but such inventories can be made only after far more intensive investigations than have so far been practicable. Tipi-ring sites occur in great numbers in the Tiber area. They seem to be most common on the plateau-like bluff tops overlooking the river valley (pl. 7, fig. 2), and are less frequently met with on the terraces. Associated with some of these are small piles of stone, usually circular in outline and about 5 feet in diameter. Tests in two of these piles disclosed small unidentifiable fragments of bone beneath them, but no satisfactory evidence that the underlying soil had ever been disturbed. Their purpose remains conjectural. It is possible that the tipi rings and associated remains are relatively late ; very little stonework occurs with them and the stones composing the circles usually lie on or very near the ground surface. No metal or glass objects were noted on these sites. No pottery-bearing sites were noted by the River Basin Surveys field party at Tiber; and such remains appear to be absent generally from the area. A site of very considerable interest is a bison kill, 24LT22, located about 5 miles from the proposed dam site and somewhat above full pool level. It is in a shallow forked draw, where the animals were apparently ambushed, perhaps with the aid of log or brush barricades 36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE. Ea or partial enclosures. Extensive pitting has been done by relic col- lectors. In these pits and in test excavations by the River Basin Surveys party, the soil contains partly burnt and decomposed bison hair, hooves, and bones, charcoal, ash, and fire-broken stones. The tests also showed that this material occurs in two, three, or more levels separated and capped by thin strata of culturally barren soil. The observed thickness of deposit varies from 1 foot upward, with the deepest bone layer noted at approximately 30 inches; greater depths have been reported by local collectors. Numerous small projectile points, mostly triangular in form with a single pair of side notches, were recovered. There is some evidence that significant type differ- ences may exist between the points from various parts of the kill area and also between those from different levels. Local reports that “post holes” have been noted from time to time, if correct, suggest the use of corrals or barricades. PALEONTOLOGY The Missouri River Basin has long been noted among scientists for the wealth and variety of its paleontological resources. Systematic collecting of the remains of extinct animals and plants began with Hayden’s geological survey of the Nebraska Territory in the middle of the nineteenth century. In the past 90 years, field parties from many museums and institutions have been successfully at work in numerous widely scattered localities. It has been estimated that three- fourths of the fossil materials in several of the larger museums of the nation were gathered in the Missouri River watershed. The extraordinary richness of the Basin in this respect is due to its vast extent, its varied topography, and the long span of geologic time revealed thereby. At one place or another, the processes of uplift, deposition, and erosion have brought to light exposures of nearly all the geologic periods from Paleozoic times to the Recent. In these exposures are to be found a great variety of plant fossils, as well as vertebrate remains ranging from fishes through reptiles, including such spectacular giants as the Brontosaurus and the Dip- lodocus, to birds and mammals. In order to meet its commitments to the National Park Service and thé construction agencies with respect to the possible effect of water-control projects on paleontological resources, the River Basin Surveys added a paleontologist to its professional staff early in 1947. In this capacity, Dr. T. E. White arrived at the Lincoln headquarters of the Missouri Valley project on April 29 and prepared for im- mediate field work. NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 37 From May 2 to May 8 Dr. White visited seven proposed reservoir areas in the Lower Platte Basin in north-central Nebraska, including Amherst, Brewster, Buffalo Creek, Cairo, Ericson, Mullen, and Rockville. Fossil remains were found only in Mullen, and these were reworked material of no scientific interest. On May 13, a recon- naissance of the Smoky Hill and Republican Basins in southwestern Nebraska, northern Kansas, and northeastern Colorado was begun ; by June 6, 19 reservoirs had been visited. These included Beaver City, Buffalo Creek, Culbertson, Enders, Harlan County, Medicine Creek, Rock Creek, and Red Willow Nos. 1 and 2, in Nebraska; Cedar Bluff, Glen Elder, Kanopolis, Kirwin, Norton, Pioneer, Webster, and Wilson, in Kansas; and Bonny and Wray, in Colorado. On the basis of material seen and the exposures available, more ex- tended investigations were recommended for Beaver City, Bonny, Cedar Bluff, Enders, Harlan County, Medicine Creek, and Red Willow Nos. 1 and 2. From June 13 to June 28 Dr. White examined the proposed Glendo Reservoir area in the North Platte Basin in Wyoming; Angostura and Deerfield in South Dakota, and Edgemont and Keyhole in Wyo- ming, all in the Cheyenne River Basin; Bixby and Green Grass, in the Moreau River Basin, and Blue Horse and Shadehill, in the Grand River Basin, all in South Dakota; Dickinson and Heart Butte, in the Heart River Basin, Cannon Ball in the Cannonball Basin, and Broncho, on the Knife River, all in North Dakota. Further work was recommended at Edgemont and Blue Horse. For most of the others, suitable geologic exposures were lacking, or the formations involved are much better exposed outside the proposed reservoir areas. Between July 12 and 31 the following proposed reservoir projects in Wyoming and Montana were visited: Kortes, Onion Flat, Soral Creek, Du Noir, Boysen, Anchor, Badwater, Oregon Basin, Kane, Red Gulch, South Fork, Bull Creek, Triangle Park, Lake Solitude, Smith, Willow Park, and Middle Fork, in Wyoming; Little Horn, in Montana; and Yellowtail and Moorhead, each of which lies partially in both States. Anchor, Badwater, Boysen, Middle Fork, and Moor- head should be revisited for more intensive investigation. From August 9 to September 25 the paleontological survey was extended to the following localities: Crosby, Jamestown, Sheyenne, and Souris, in North Dakota; Medicine Lake, Sunlight, Thief Creek, Hunter Mountain, Taylor, Bridger, Mission, Sweetgrass, Antelope, Newland, Stanford, Hobson, Ross, Snowy, Tiber, Wilson, Nilan, Wells, Canyon Ferry, Terry, Whitetail, Apex, Kelley, Clark Canyon, Brenner, and Landon, in Montana. Of these, Medicine Lake, Sun- 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III light, Mission, Brenner, Canyon Ferry, Tiber, Ross, Hobson, and Snowy appeared sufficiently promising to be recommended for further attention. This trip was interrupted between August 20 and September 11, during which period Dr. White and his assistant excavated the skull and other skeletal parts of a dinosaur at Middle Fork Reservoir in Johnson County, Wyo. The dinosaur is from the Upper Jurassic Morrison formation, and has been identified as Diplodocus. On October 7, Dr. White left Lincoln again to conduct intensive survey and fossil-collecting at Boysen. Enroute he examined paleonto- logical collections at the Zeitner Museum, Mission, S. Dak., and made preliminary inspections of the proposed Philip and Rocky Ford Reservoir areas, respectively located on Bad River in Haakon County, S. Dak., and on the White River in Washington County, S. Dak. The field work at Boysen terminated on November 7. Most of the collecting during this period was in the Eocene Wind River forma- tion in the Cottonwood Creek drainage on the west side of Big Horn River. Among the specimens obtained the following may be noted, all represented by incomplete remains: a lemuroid (Pelycodus), a primitive tapir (Heptodon), an extinct mammal (Coryphodon), and the carapace (pl. 6, fig. 2) of a soft-shelled turtle (Amyda). Prospects in this locality are believed to be very promising, and plans are for further work next spring. This would involve a more intensive reexamination of the Cottonwood Creek locale, as well as investigation of other known fossil localities in the basin area. In summary, paleontological reconnaissance was carried out at 94 proposed reservoir areas between May 2 and November 7, 1947. Twelve of these are in intrusive granite and contain nothing of interest to paleontology. Seven are in geologic strata from which vertebrate fossils are at present unknown. At 25, the reconnaissance showed that fossil deposits will be directly affected and recommendations have been made for further studies if and when construction work is initi- ated. At the remaining areas, the vegetation cover precluded profit- able search for fossils on the existing land surface, though the possi- bility remains that construction activities may in some places uncover worth-while materials. As might be expected, the preliminary nature of most of the pale- ontological work to date has resulted in acquisition of relatively little exhibit or study material. Outstanding finds have been noted else- where in this report. A highly worth-while working knowledge of the units visited, of their geologic setting, and their paleontologic potenti- alities has been gained. It seems certain that intensification of this NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 39 work, with emphasis naturally on the more promising and urgent units, will add much significant new information to our knowledge of the paleontological resources of the Missouri River Basin. FIELD WORK BY COOPERATING AGENCIES It is gratifying to note that a number of State-supported agencies have oriented their own archeological and paleontological field research programs in order to assist in the scientific salvage operations in the Missouri River Basin. In some instances this has necessitated aban- donment, or at least indefinite suspension, of plans previously formu- lated. Such agencies, where they have indicated a wish to participate in salvage work, have been granted letters from the River Basin Sur- veys head office at Washington, authorizing them to undertake re- searches in specified localities or at particular sites. This cooperation is on a voluntary basis and involves no distribution of Federal funds to the State agencies. Cooperating institutions and organizations provide reports on their operations at regular intervals and otherwise as requested, and summaries of these reports are transmitted by the Lincoln field office to the regional office of the National Park Service for such disposition and distribution as is necessary or advisable. This section of the present report is based largely on data furnished by the agencies involved. Some of the sites noted have been visited by River Basin Surveys personnel while the State-supported work was under way. MISSOURI Because of more pressing construction schedules elsewhere, the Missouri River Basin Survey has undertaken as yet no field investiga- tions in Missouri, where a number of water-control projects have been proposed by the Corps of Engineers. The University of Missouri, a cooperating agency, has been active in this region, with primary emphasis on high-priority projects lying outside the Missouri River watershed. Working with the State university are two other groups: the Missouri Resources Museum, at Jefferson City, and the Missouri Archeological Society. Active direction of the field work has been in the hands of Carl Chapman, of the university. The archeological salvage program in this State has been, and is, of particular interest in that it demonstrates the usefulness of a coordi- nated effort by professionals and nonprofessionals. The State uni- versity has acted as a directing agent and a clearinghouse for informa- tion and planning. Technical advice and suggestions, and in many 4 40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III cases direct assistance in the field, have been offered to members and local chapters’ of the Missouri Archeological Society; the latter, in turn, have reported to the university the results of their field in- vestigations. As requested, these data have been placed at the disposal of the River Basin Surveys to assist in preparation of reports and recommendations to the National Park Service and the construction agency. Of the several Corps of Engineers projects proposed for the Missouri River watershed in Missouri only Pomme de Terre Reser- voir was surveyed in 1947. This is located on the Pomme de Terre River, a southerly tributary of the Osage. A joint survey here by the University of Missouri and the Ozarks Chapter, Missouri Archeo- logical Society, disclosed the location of 25 sites of archeological interest. These include open camp sites, stone cairns or graves, and caves or rock shelters containing evidence of former human occupa- tion. Pottery was found at only one site; its apparent absence from the others, together with the types of stone artifacts noted, suggests that a rather lengthy period of habitation by prepottery peoples may be represented. It is noteworthy that one of the earliest records in North America of human artifacts associated with bones of extinct Pleistocene mammals was made by Albert Koch in 1840, only a few miles downstream from the proposed Pomme de Terre dam site. Thus, despite the brief time devoted to survey in 1947 in this reservoir area, it seems clear that antiquities of considerable promise will be affected and that a close watch will have to be maintained while con- struction is in progress. In addition to Pomme de Terre, investigations were carried on in 1947 at Joanna Reservoir, on Salt River in northeastern Missouri ; at Clearwater Reservoir, on the Merrimac River in eastern Missouri, and at Bull Shoals Reservoir, now under construction on White River in the southwestern part of the State. Numerous sites have been located and recorded, and at Bull Shoals a camp site and a rock shelter, both stratified, were excavated. As elsewhere in the Basin, so in Missouri it is evident that a wide range in time and a variety of remains from several different periods will be directly affected by the water-control program. It is the intention of the agencies active in this area to continue the salvage operations. NEBRASKA In Nebraska, the Smithsonian Institution has cooperative agree- ments for archeological work with the Nebraska State Historical Society and the Laboratory of Anthropology, University of Nebraska, NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 4!I and for paleontological work with the University of Nebraska State Museum. All these State agencies conducted field work, in varying degree, during 1947. The Nebraska State Historical Society carried on archeological excavations at Medicine Creek Reservoir from July 25 to September 10. This work was under the personal supervision of A. T. Hill, director of the museum, who was assisted by students and local work- men. Excavations were made chiefly at two sites near the proposed location of the dam, with test work at several other points in the future reservoir area. Near the west end of the proposed dam axis, on site 25FT16, the floors of two rectangular, semisubterranean earth lodges were uncovered. The floors were approximately 30 feet long, slightly less in width, and lay at a maximum depth of about 18 inches underground. Each had a central firepit, post molds showing four primary roof supports and others showing straight walls, rounded corners, and a covered entrance passage opening to the south. From the floors and the fill immediately above, and from caches within and nearby. middens outside the structures, were recovered pottery frag- ments, chipped- and ground-stone work, shell and bone artifacts, charred corn, and refuse animal bone. The materials can probably be safely assigned to the prehistoric semihorticultural Upper Republi- can horizon. Just above the mouth of Lime Creek, some 2 miles northwest of the proposed dam site, a third house floor was opened at site 25FT28. Except in its somewhat smaller size, this differed in no important particular from those at site 25FT16. Several restorable vessels were found on the floor, along with other artifacts and miscellaneous materials, and these again indicate an Upper Republican complex, though with some variations in details. Limited tests were made at other sites in the locality. The materials collected and the basic field data were placed at the disposal of the River Basin Surveys office at Lincoln for processing and study. As indicated elsewhere in this report, archeological excavations were continued in the Medicine Creek area by the River Basin Surveys after cessation of the work by the Nebraska State Historical Society. A one-day reconnaissance of the proposed Bellwood and Shell Creek Reservoirs was made by Dr. J. L. Champe, University of Nebraska Laboratory of Anthropology, and Mr. Hill, of the Historical Society. Subsequently, rapid preliminary reconnaissance was made of the proposed Clearwater, Loretto, Davis Creek, Cushing, and Plum Creek Reservoirs in the Lower Platte Basin, by Dr. Champe in company with Dr. T. E, White, paleontologist for the River Basin Surveys. 42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES EieL As special consultant for the River Basin Surveys, Dr. Champe also spent approximately 1 week as observer at State-sponsored excava- tions on Lime Creek in the Medicine Creek Reservoir area. The University of Nebraska State Museum carried on extensive investigations on Lime Creek, a small westerly tributary of Medicine Creek lying within the area to be flooded by the proposed reservoir. One fossil quarry and three sites where archeological materials are reported in association with fossil bones were worked. These investi- gations were under the supervision of Dr. C. B. Schultz and W. D. Frankforter. The fossil quarry is attributed to the very late Pliocene period. Several new forms of extinct mammals are reported to have come from it, including the skull of a well-preserved saber-tooth cat at first pronounced by those in charge of the work to be a marsupial of South American type. Additional light is promised on paleontological prob- lems of the Pliocene-Pleistocene transitional period, when detailed analysis of the findings here will have been made. Of interest to archeologists no less than to paleontologists are the finds at three sites situated in the basal portions of a terrace identified by the University Museum investigators as Republican River Terrace 2 and assigned a late Pleistocene dating. At the principal locality, site 25 T41, evidences of former human activity occur in a dark-gray stratum 474 feet below the terrace surface (pl. 8, fig. 1). This pre- sumably represents the valley floor at the time of human occupation. The overburden consists of silts and loess, the upper 17 feet of which have been correlated tentatively with the Bignell loess, thought to have been deposited during the Mankato stage of the Wisconsin glaciation. Points are said to have been found in situ; in addition, there were leaf- shaped and other blades, end scrapers, knives, fragments of a grooved sandstone “shaft-smoother,” numerous flakes, spalls, cores, and mis- cellaneous rejectage. Worked bone and antler are also reported. These were associated with bones of some 17 mammalian forms, as well as those of reptiles, birds, and amphibians. Preliminary observa- tions “suggest distinct differences between several of the fossil and modern forms, but positive identification must wait for further preparation and comparisons.” The full report on this important site is awaited with keen interest. Typologically, few of the artifacts seem to differ markedly from many of those found in later pottery-bearing horizons of the region. Among the individuals who have actually visited the site, there are rather marked discrepancies in interpretation as regards the apparent age and the relationships of the archeological remains. It seems obvious NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 43 that a manifestation with the potential importance indicated here should be thoroughly investigated by trained specialists in archeology, as well as by geologists, paleontologists, and soils experts. A com- bined attack with all interested disciplines represented would doubt- less remove many of the doubts and uncertainties that now surround the findings, and would enable the site or sites to be placed in their proper geological and archeological setting. SOUTH DAKOTA In South Dakota, there was no formal cooperative arrangement between Federal and local agencies for scientific salvage work in 1947. Archeological investigations were carried on by State-supported agencies, however, and most of this work was in areas that will be affected by the Federal water-control program. A summary of the findings is included therefore in this report. Field work from June 10 to September I was sponsored jointly by the University of South Dakota Museum, Vermillion, and the newly created South Dakota Archaeological Commission, Pierre. Funds for this work came partly from the State, and partly by private sub- scription. E. E. Meleen was in charge of the field operations, with W. H. Over acting in a general supervising capacity. From June Io to June 30, excavations were carried on at the LaRoche site, about 25 miles southeast of Pierre on the right bank of the Missouri River in southeastern Stanley County. Situated on a low terrace and marked by inconspicuous refuse-littered mounds, this site is expected to be inundated by the proposed Big Bend Reser- voir. Two circular lodge sites were opened, each characterized by four center post molds, a central firepit, and a formerly covered entryway opening toward the southeast. Potsherds, and objects of chipped and ground stone, bone, horn, shell, and catlinite were re- covered. Charred corncobs were quite common, predominantly of the 10-row variety. The pottery shows many similarities to that from protohistoric Pawnee village sites on the Loup River in central Ne- braska, as well as to that from the upper levels of the Scalp Creek village site in Gregory County, S. Dak. From July 1 to July 23, work was conducted at the Somers site, on a high bluff about 2 miles northwest of the LaRoche site, in Stanley County. Here numerous house pits are still visible, although the former village living level is buried beneath 54 inches or more of fine wind-blown dust. Limited manpower combined with the heavy overburden prevented more than a light sampling of the site. Two 44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III contiguous walls of one lodge measuring 25 feet in width by 36 feet in length were uncovered, and a portion of one wall in a second. Both indicated a rectangular pit-house form, apparently with straight walls and slightly rounded corners. Well-preserved sections of cedar house posts, from which it may be possible ultimately to determine the date of occupancy, were recovered. Pottery fragments were relatively plentiful; most of the sherds have cord-roughened surfaces, and the general impression is one of close relationships to the prehistoric Upper Republican complex of Nebraska and Kansas. Other items found include bison-scapula hoes and charred cobs indicating a horticultural subsistence basis; bone fishhooks, awls, bone and shell disk beads, flakers, small, well-made notched and plain triangular projectile points, scrapers, leaf-shaped quartzite knives, and drills. Many of these items are also reminiscent of the Upper Republican horizon farther south, and a basic relationship is implied, even though the house types suggest possible eastern connections or influences. It seems safe to conclude that the Somers site represents an earlier time period and a different cultural complex from that manifested at LaRoche. At neither was there evidence of contact with Europeans. Following work at the Somers site, operations were transferred to the Thomas Riggs site, on the left bank of the Missouri above Pierre in Hughes County. This site, which will be flooded by Oahe Dam, had been investigated briefly in 1940 by the University of South Dakota Museum and Works Progress Administration. In 1947, the excavation of a large semisubterranean lodge site begun in 1940 was completed. Rectangular in shape, this house was outlined by a double row of post molds and charred posts along each of the two longer sides and a single large post (12 inches) in the center at the rear end (east) and two at the front, one on each side of the entryway or ramp leading down onto the lodge floor. This structure was found to be 65 feet long by 36 feet wide. From the limited amount of pottery and other material gathered from the site, connections with the Mandan are suggested. These, however, are highly tentative; like the data from LaRoche and Somers, further information and more detailed analysis than has so far been possible is needed before wider relation- ships of the peoples represented can be suggested. NORTH DAKOTA In North Dakota, an archeological field session was sponsored jointly by the Department of Sociology and Anthropology, University of North Dakota, and the North Dakota Historical Society. A party NO. ‘2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 45 of six students, under the leadership of Dr. Gordon W. Hewes, de- voted 6 weeks to the investigation of sites in proposed reservoir areas on Heart River and on the Missouri. From June 25 to July 1 this expedition worked in the Heart Butte Reservoir area in Grant County, south of Glen Ullin. Limited exca- vations were made in a pottery-bearing deposit (32GT1I) on the north bank of the Heart River, about 3 miles upstream from the dam site. Here, in a stratum reaching a thickness of 2 to 3 feet or more, were found traces of former occupation by a group of bison-hunting, semisedentary people, whose pottery tradition was quite similar to that of the Mandan and Hidatsa. Objects of ground stone, bone, horn, shell, wood, and other materials were absent or very scarce; chipped- stone work included end scrapers, drills, blades, and point fragments. No evidence of earth-lodge habitations, of agriculture, or of contact with white people were found. From the abundance of their bones, bison seem to have been the chief food item, but there was also con- siderable use of river mollusks. It is suggested that this site, pre- viously recommended for excavation by a reconnaissance party of the River Basin Surveys, may represent a camping place occupied seasonally by hunting parties of the Mandan or Hidatsa, whose vil- lages lay 50 or 60 miles to the east on the Missouri River. Surveys made concurrently with the excavations located a small rock shelter (32GT5) near the dam site. From the very thin floor deposit came a few pieces of chipped chalcedony and fragments of a single pottery vessel of late Mandan-Hidatsa type. Fallen slabs in front of the shelter, underlaid by cultural debris, suggest that addi- tional data may be buried beneath the collapsed front roof of a once deeper shelter. Upstream from the camp site was found a rather extensive deposit of bison bones, evidently representing the debris of a hunting drive or “kill.” This deposit, now buried by 12 to 13 feet of overburden, was exposed for nearly 500 feet along the river bank; no artifacts were noted. On higher ground, above the future reservoir level, was found a chalcedony quarry; scattered spalls and a few chipped implements testify to the use of the material by the Indians. From Heart River, the expedition moved to Fort Yates on the Standing Rock Indian Reservation. Excavations were undertaken in an earth-lodge village site (32514) 7 miles south of Fort Yates, on the right bank of the Missouri River. Like an undetermined number of others along the stream south of Bismarck, this site will be inun- dated eventually by the proposed Oahe Reservoir. On the basis of surface sherd collections, it had been previously ascribed by North Dakota workers to the “Archaic Mandan” horizon. Surface remains 46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III otherwise consist of a number of bowl-like depressions varying in depth up to 3 feet or more and in diameter up to approximately 50 or 60 feet. There is no evidence of a protective ditch or embankment. Preliminary excavations in several of the depressions disclosed ash-filled fireplaces and rather poorly defined floor levels, indicating the former presence of semisubterranean house structures. Test pits in various other parts of the site revealed food caches, secondarily used for refuse disposal by the Indians. Most of the 5 weeks spent at the location were devoted to examination of the feature represented by the largest surface depression. Removal of the fill within the depression showed that the structure which once stood here had been subrectangular in outline, measuring approximately 35 by 65 feet. Circular discolorations in the floor marked the positions of the four rows of posts outlining the house area. A single small pole was recovered from one side, but most of the posts seem to have been removed when the structure was aban- doned. From the position of the post molds, it is suggested that there was some sort of long central hall in the structure. The arrangement of one large and two small fireplaces, the scarcity of household refuse, the presence of three piles of bison bones (mostly unburned skull parts and horn cores) on the floor, and the exceptional size of the structure as compared to other depressions on the site, suggest that it may have been used primarily for ceremonial or other special pur- poses. Two large pots, broken but restorable, were recovered. Agri- culture can be inferred from discovery of several bison-scapula hoes and a fragment of carbonized maize cob, but bison appear to have been a major dietary item. No burials were encountered, nor was there any evidence of trade contacts with white men. Materials found during excavation include some bone tools, incised bone ornaments, two circular shell beads, numerous small end scrapers, a few side scrapers, projectile points, knives, choppers, scapula hoes, “squash knives,” bison-rib beamers, a fragment of carbonized corncob, and much animal- and bird-bone refuse. Pottery was not abundant on the site, but from some of the refuse pits and the fill of the structure excavated came a fair sample. A rather complex problem is presented. In addition to types that seem clearly in the tradition which culminated in the historic Mandan- Hidatsa wares farther upriver, there are numerous fragments that suggest borrowings from, or more direct relationships with, Upper Republican, Mill Creek, Cambria, and perhaps western or “Prairie” Hopewellian traditions. The site evidently falls somewhere in the prehistoric period of development of Upper Missouri village Indjan NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 47 culture that culminated in the historic Mandan-Hidatsa complex observed farther to the north in and after the middle of the eighteenth century. Tentatively, it is suggested that the time period represented may be in the century between 1350 and 1450. An area of several acres extent on Four-Mile Creek, about 2 miles from the Missouri, was found to be littered with bison bones eroding from an indurated clay stratum. Several flints that may represent crude tools were found in the same area, but their association with the bones is uncertain. There is a possibility that a prepottery hunt- ing culture is represented here, but additional research is necessary to determine the true facts in the matter. CONCLUSIONS From the reconnaissance, intensive surveys, and limited excavations carried on through 1947, it is manifestly impossible to reach any fixed or definitive conclusions regarding broader aspects of the aboriginal history of the Missouri River Basin. It can be stated truthfully that the River Basin Surveys work to date has served mainly to indicate the magnitude and complexity of the problems involved, with some rather promising leads as to where partial answers to some of the problems may profitably be sought. Throughout the body of this brief report I have occasionally ventured opinions, tentative conclusions, and suggestions. In concluding this summary, therefore, I shall try to indicate only some of the highlights of work so far and to suggest certain additional avenues of approach to further information. Primary emphasis will be on the work of the River Basin Surveys, with which I am most intimately familiar. Noteworthy is the fact that the sampling technique necessarily imposed on the River Basin Surveys program thus far is producing an impressive mass of information from widely scattered localities in the Missouri River watershed. Operating through 1947 mainly on the first, or reconnaissance, phase of the salvage work, we now have spot-check data on the archeological resources of approximately 50 localities, each of restricted areal extent. In each locality where remains have been found, there are indications of aboriginal occupa- tions of diverse character; and frequently these, when appraised in the light of what is known for the general area, also give clues to the succession of native occupations. When these various now sepa- rate bits of information will have been properly classified and fitted into their appropriate place in the over-all picture, definitive conclu- sions may be expected to result. 48. . SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III In the Wyoming-Montana area, the surveys of 1946-47 have shown that a great number of sites exist and that the water-control program will adversely affect many of them. As has long been inferred on historic grounds, this was primarily a region of simple hunting and gathering economies, with no native horticulture except sporadically along the extreme eastern margin. Pottery occurs sparingly here and there—as at Glendo and Boysen, in Wyoming; and perhaps somewhat more plentifully along the valley of the Yellowstone in Montana. The so-called tipi rings, whose true purpose and signifi- cance are still obscure, are abundant and apparently highly character- istic. They occur in limited numbers in northern Colorado and ex- treme western Nebraska, and more frequently in the Dakotas east- ward approximately to the Missouri from Fort Randall northward, but the greatest number of such sites seem to lie in Wyoming, Montana, and northward. It is possible that they correlate with a relatively late hunting occupation, perhaps partly at least involving Shoshonean peoples. Camp sites marked by clusters of fire-cracked stones and refuse, but without tipi rings or other evidences of structures, are also common, particularly in the western portions of the area. There is some reason to believe that many of these hearth sites are of some antiquity, since they appear to be weathering out of cut banks at vary- ing depths below the present surface. Whether they represent an early Shoshonean occupation, or are pre-Shoshonean in time, or both, is uncertain. Determination of cultural succession in this region is not easy, because of the simple nature of most of the site complexes and the comparatively low material yield from most occupational sites. That a series of peoples have successively inhabited the area, and that they have carried different cultural equipment, is already demonstrable, however, and there can be little doubt that continued intensive investi- gation at appropriate sites will lead to clearer definition of variations now only dimly recognizable. Of several stratified sites recorded by River Basin Surveys personnel, only one has been adequately studied. This is Birdshead Cave, in the Owl Creek Range west of Boysen dam site. Here it appears, from evidence noted elsewhere in this report, that peoples with a Great Basin type of subsistence economy rather than Plains hunters dwelt in late prehistoric times. Aside from the somewhat more advanced and more abundant remains in the upper levels, the principal differences from period to period are indicated by variations in form and size of projectile points. It seems probable that these objects, varying from well-made, triangular, side-notched forms in the later period to progressively earlier corner-notched and | | NO; -2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL 49 then unnotched concave-based forms, may be of primary importance as a guide to the sequential arrangement of innumerable single-com- ponent sites in this largely potteryless region. There are several widely scattered sites that promise, if excavated, to yield important informa- tion on Early Man and related problems. At the other end of the time scale are sites, some of them beneath alluvial or aeolian deposits, such as in the Tiber Reservoir area on the Marias River, Montana, where glass beads and metal occur in association with well-defined hearths and occupation strata. Brief mention has already been made of the bison kills of this region. These localities, where masses of the animals were slaughtered by being stampeded over cliffs or steep bluffs, or were perhaps am- bushed in broken terrain, are widely scattered throughout Montana, Wyoming, and the western Dakotas. They seem to be particularly numerous along the stream valleys of western Montana, as in the Teton, Sun River, and adjacent areas. Many have been dug into by local collectors, who report the finding chiefly of great numbers of notched triangular projectile points among the bones. Some of the bone deposits in these kills or traps apparently cover thousands of square feet in extent and exhibit some depth, suggesting repeated use of the spot. Differences in form and size of projectile points are observable, and one suspects some stratigraphic variations that may be of chronological and developmental significance. It is highly de- sirable that systematic stratigraphic excavations be made in a number of these sites, to ascertain their age, length of use, and relationships to the archeological horizons represented at camp and occupational sites in the region. Very ancient, that is, paleo-Indian, remains have apparently seldom been found in these kills ; and despite the suspected recency of some, iron arrowpoints and evidence of use of firearms are very rare or absent. Rock alignments in this region, including converging rows of small boulder piles said to be sometimes associated with bison kills, also remain mostly unexplained and the period of their construction undetermined. . In the eastern portion of the Missouri River Basin, from North Dakota through South Dakota and Nebraska into northern Kansas, the River Basin Surveys have been concerned largely, but not ex- clusively, with the remains of semisedentary, pottery-making peoples. The remains of their former villages decrease in abundance and variety from east to west, with the largest villages and the climax of their cultural development shown along the banks of the Missouri and on some of its major tributaries. Some of these peoples, as already 50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Tit noted, erected burial mounds along the stream valleys of eastern North and South Dakota. In the Garrison Reservoir area north of Bismarck, N. Dak., the work of the River Basin Surveys has thrown light on the problem of what we may term the northwestern periphery of the Upper Missouri culture area. Here the fortified earth-lodge-village complex, so abundantly represented farther downstream, fades out ; tipi rings, buried camp sites, and other vestiges of a less settled mode of life become more characteristic. Noteworthy is the stratification of cultures suggested at several sites: heavy cord-roughened sherds and large-stemmed projectile points underlying simple-stamped pottery and small triangular or side-notched points. It would seem from this that early Woodland manifestations preceded the Mandan-Hidatsa village complex in the northern Great Plains, as they do the earth- lodge-village dwellers in the Central Plains. At the moment, it does not appear that these Woodland-like remains occur to any great extent west of the Missouri River in the Dakotas, though further field work may change this impression. Sherds bearing well-made dentate stamp impressions and reminiscent of certain Illinois Valley specimens have been found at least as far west as Medicine Lake in northeastern Montana north of the Missouri. Check-stamped, simple-stamped and other sherds have been noted in the Williston district of western North Dakota. In general, what are perhaps the earlier ceramic sites in the region appear to show affiliations with the Woodland horizon of Min- nesota whereas the later material is suggestive rather of village in- fluences from the Mandan-Hidatsa area. At Baldhill Reservoir, where the University of North Dakota plans investigations in the summer of 1948, important mound and village sites are present. Their strategic location with reference to east-to- west movements in late prehistoric and protohistoric times, and the rather strong evidences of an interplay of Woodland cultures from the Minnesota area with more distinctively Plains complexes, lends urgency to the problem of salvage and systematic study. Farther down the mainstem in South Dakota, at Fort Randall Reservoir, further important information on the problem of Central and Northern Plains interrelationships was gathered in 1947. At several earth-lodge village sites there are strong evidences of an Upper Republican-like complex, with ceramic traits strikingly like those from the Republican Valley in southern Nebraska. Other sites ex- hibit pottery wares, house types, and other elements strongly remi- niscent of the protohistoric Lower Loup complex in east-central Nebraska, suggesting that the Arikara may have moved northward NO. 2 MISSOURI VALLEY DEVELOPMENT PROGRAM—WEDEL SI at an earlier date than is commonly supposed. Still earlier, presum- ably, are several small, scattered mound groups; one of these near Wheeler Bridge included mounds with log-covered burial pits con- taining disarticulated burials, perforated long bones, and other ele- ments heretofore unreported from the Great Plains or the upper Missouri. From the limited excavations so far made in this section, it is evident that the late prehistoric and protohistoric archeology will be rather complicated, as may be anticipated from what we know of tribal movements into and through the locality from several direc- tions. The presence of stratified sites leads to the conviction that an orderly arrangement of cultural events will be feasible when enough systematic work will have been done. Still farther south, in the Lower Platte Basin and adjacent westerly subdrainages of the Missouri, promising data have been accumulating on several inadequately known archeological complexes of the Central Plains. Of particular interest is a series of sites represented at Harlan County, Ericson, and Mullen Reservoir areas. The pottery and some other remains show similarities to sites at Glen Elder and on White Rock Creek, in northeastern Kansas, and also to certain sites in northeastern Nebraska. The complex has not yet received careful study, but it seems to offer some promise of throwing light on such late prehistoric or protohistoric tribal movements as one would suspect for the Ponca, Kansa, and perhaps other Siouan tribes, or possibly for certain northern Caddoan peoples. Probably belonging to the same general time period, but of different tribal affiliation, are the several Dismal River culture sites that have come to light, one in Harlan County and several in Mullen Reservoir area. This complex is a little better known than the preceding one, and seems to be attributable to a late seventeenth- or early eighteenth-century Apache or possibly Comanche occupation. Clarification of the temporal and cultural position of both these complexes awaits further field work and careful laboratory analysis. For such earlier Central Plains cultures as Upper Republican and Woodland, the 1947 field work of the River Basin Surveys has also gathered significant distributional and other data. It is becoming increasingly evident that there are variations in both which probably have developmental significance, but only further excavation of sites on a comprehensive scale can enable us accurately to define and interpret these variations. It is worth noting that the 1946 excavation by the River Basin Surveys of an ossuary on Prairie Dog Creek, near the upper end of Harlan County Reservoir, promises to compel re- vision of current concepts regarding burial practices of Woodland 52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III and Upper Republican peoples, and also to contribute materially to clarification of the problem of interrelationships of these two archeo- logical horizons. No evidence of pre-Columbian puebloan contacts with Central Plains peoples has been found. In retrospect, it is scarcely necessary to reiterate that the surveys to date have gathered in a great quantity of useful archeological and human ecological information for many sections of the Missouri River Basin that will be directly affected by the water-control pro- gram. By comparison with the returns that might be realized through detailed excavations following up the leads now at hand, the salvage task has just begun. It has barely touched some of the potentially richest sections of the Missouri Valley. As Cooper has aptly ob- served in his preliminary appraisal of the archeology of Fort Randall Reservoir : Anthropologists have for years recognized the upper Missouri as one of the richest and most promising archeological areas in North America. In historic times, an important part of the fur trade between whites and Indians was carried on at the great stockaded towns of the Mandan, Arikara, and their neighbors on the mainstem in South and North Dakota. These towns represented a com- paratively advanced stage of native civilization, basically of an agricultural character, and were inhabited by what were apparently only the last of a series of people who at various times and perhaps from several directions occupied the region. Progressively simpler and less advanced peoples, who relied to a great extent on hunting, seem to have preceded the Indians first seen in the region by white men. The steps by which a highly specialized corn-bean-squash economy, adapted to the rather trying environment of the upper Missouri, evolved out of the native agricultural economies to south and east, remain to be worked out. There are suggestions that the prehistoric farmers of the area may have been beset, perhaps even displaced, from time to time, by drought, floods, and other vagaries of nature. A long and complex story of man’s struggle with his environment, without the technological advantages of the white man today, thus awaits closer scrutiny. The camp sites, villages, towns, and burial places of the region represent the documents from which this story must be assembled. It is this story, rather than the mere accuraulation of specimens and compiling of lists of site characteristics, which is envisaged in the archeological research program proposed. In varying degree and with local qualifications, the above evalua- tion applies to all archeological remains subject to damage or destruc- tion by the Federal water-control program in the Missouri River Basin. The challenge is obvious. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 2, Pl. 1 1. DISARTICULATED OR BUNDLE BURIALS IN GRAVE PIT BENEATH MOUND Traces of log cover are visible on pit edges. Site 39CH4, near Wheeler Bridge, Fort Randall Reservoir, S. Dak. Neg. 39CH4-48. 2. BUNDLE BURIAL ON FLOOR OF GRAVE PIT Note perforations in leg and arm bones. Site 39CH4, near Wheeler Bridge, Fort Randall Reservoir, S. Dak. Neg. 39CH4-40. AN M ANEOUS COLLECTIONS Vol. 131, No. 2, Pl. 2 uw 1. EXCAVATED FLOOR OF SEMISUBTERRANEAN EARTH LODGE, POSSIBLY ARIKARA Firepit in center, four large central and numerous small outer post holes, and vestibule entrance opening toward the southeast. Oldham Site (39CH7), Fort Randall Reservoir, S. Dak. Neg. 39CH7-9. 2. BURIED ARTIFACT STRATUM IN MISSOURI RIVER TERRACE, NEAR CHAMBERLAIN, S. DAK. Site 39BR11, Fort Randall Reservoir, S. Dak. Neg. 39BRII-5. SMITHSONIAN MISCELLANEOUS COLLECTIONS Volt Lil Nok2, Pls 3 1. TESTING OCCUPATIONAL STRATUM BENEATH 5 FEET OF OVERBURDEN Site 32MZ12, McKenzie County, Garrison Reservoir, N. Dak. Neg. 32MZ12-3. 2. BATTLE BUTTE, 6 MILES SOUTH OF EMMETT, N. DAK. The flat summit is littered with flints, sherds, and other evidence of Indian occupancy; Missouri River at left. Site 32MLo0, McLean County, Garrison Reservoir, N. Dak. Neg. 32MLo-1. ITI / MI _LLANEOU COLLECTIONS Vol. alata, No. 2, Pl. 4 1. POST SECTION EXPOSED BY MISSOURI RIVER CUTTING INTO REMAINS OF STOCKADE SURROUNDING ROCK VILLAGE (HIDATSA?) NEAR EXPANSION TOWNSITE, N. DAK. Site 32ME15, Mercer County, Garrison Reservoir. Neg. 32ME15-3. 2. STONE-HEARTH CAMP SITE AMONG THE DUNES ON TUFF CREEK Site 48F R23, Boysen Reservoir, Wyo. Neg. 48F R23-2. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 2, Pl. 5 1. PETROGLYPHS OF UNKNOWN AGE Site 48F R13, Boysen Reservoir, Wyo. Neg. 48FR13-1. 2. EXCAVATION OF BIRDSHEAD CAVE Site 48F R54, in the Owl Creek Range, Boysen Reservoir, Wyo. Neg. 48FR54-10. NEOUS COLLECTIONS Vole Lig) INote27 I Piss6 1. LARGE ROCK SHELTER CONTAINING CULTURAL DEPOSITS ATA DEPTH OF 2 TO 3 FEET, COVERED BY STERILE DEPOSITS Site 48PA24, Oregon Basin, Wyo. Neg. 48PA24-1. aor ‘eh 2. DR. WHITE EXCAVATING THE CARAPACE OF A LARGE SOFT-SHELLED TURTLE (AMYDA) his is the first well-preserved specimen of the kind ever taken from the Wind River formation, Boysen Reservoir, Wyo. Neg. 48FR65-1. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol; 112, No. 2; (Pl.e7 1. FIELD PROCESSING OF SPECIMENS AT THE RIVER BASIN SURVEYS EXPEDITION CAMP AT GLENDO RESERVOIR, WYO. Neg. 48PLoo-6. 2. TIF! RINGS ON BLUFFS OVERLOOKING THE MARIAS RIVER, TIBER RESERVOIR, MONT. Neg. 24TL12-1. COLLECTIONS Vol. 111, No. 2, Pl. 8 AIT At LANEOUS 1. DEEPLY BURIED SITE ON LIME CREEK Figures at lower right indicate artifact- and bone-yielding stratum. Site 25FT41, Medicine Creek Reservoir, Nebr. Excavations by University of Nebraska State Museum. Neg. 25FT 41-1. 2. SHELL-BEAD-ENTWINED SKELETON OF ADOLESCENT FROM PREHISTORIC BURIAL PIT Site 14PH4, Harlan County Reservoir, Nebr. Neg. 14PH4-97. (Laboratory photograph. ) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 3 CAalcott Fund FURTHER NEW CAMBRIAN Be eve ROP HON l GASTROPODS (Witu ONE PLATE) BY J. BROOKES KNIGHT Research Associate in Paleontology U. S. National Museum (PusiicaTion 3951) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION DECEMBER 24, 1948 The Lord Baltimore Press BALTIMORE, MD., U. 8 A. Walcott Fund FURTHER NEW CAMBRIAN BELLEROPHONT GASTROPODS By J. BROOKES KNIGHT Research Associate in Paleontology, U. S. National Museum (WitH ONE PLATE) As the result of efforts to probe as deeply as possible into the origins of prosobranch gastropods I published some months ago a paper describing four genera of Cambrian bellerophonts (Knight, 1947). Since that time I have met with specimens of two more Cambrian species of bellerophonts, each seemingly representing still another genus heretofore unknown. Both of these species and genera are of great interest. One, Strepsodiscus major, n. gen., n. sp., appears to be the earliest bellero- phont yet known except the questionable bellerophont Coreospira Saito. Its discuslike form and deep V-shaped sinus without a slit reminds one somewhat of the Ordovician genus Cyrtodiscus Perner but the final half-whorl is out of contact with the others and the coil is slightly asymmetrical. Oriented as a bellerophont, the right side of some specimens protrudes slightly; oriented with this right side as “upward,” the coiling would be called sinistral, a feature that may have interesting implications that will be discussed elsewhere. Chalarostrepsis praecursor, n. gen., n. sp., resembles Strepsodiscus major superficially in its discuslike shape and its free final half-whorl, but it is entirely symmetrical and has a deep slit, the earliest known appearance of a slit on any gastropod. It resembles even more significantly Tropidodiscus Meek and Worthen (Ordovician—Devo- nian) since both have the discuslike shape and the deep slit, but in Tropidodiscus the shell is closely coiled and even involute. I regard Chalarostrepsis as probably the immediate forerunner of Tropidodiscus. With these two, we now know six genera of quite certain Cambrian bellerophonts and two more that are probably primitive bellerophonts. Coreospira occurs just above or below the boundary of the Lower and Middle Cambrian. Strepsodiscus, new, and Cycloholcus Knight seem to be of Dresbachian age. Sinuella Knight appears to be of SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 3 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III late Dresbachian or early Franconian age. Anconochilus Knight is Franconian, and Chalarostrepsis new, Owenella Ulrich and Scofield, and Cloudia Knight are of Trempealeauian, seemingly in that relative order. These stages are Upper Cambrian. That each of these six genera, except Coreospira, is represented by only a single known species makes it highly probable that other species will be found that will extend the recorded ranges. It is probable too that other genera are still below the horizon. Never- theless our gradually clearing vision of Cambrian prosobranch gastro- pods and their occurrence in the time-stratigraphic sequence gives some basis for phyletic studies even though the results of such studies are naturally tentative. It is to be emphasized that all but Coreospira, which is to be regarded as extremely primitive if indeed it is a true bellerophont, have been found only in the Upper Cambrian. The Middle and Lower divisions of the Cambrian so far as we know are destitute of definite prosobranchs and, of course, opistobranchs and. other derived groups. The asymmetrical prosobranchs, as distinct from the basically symmetrical bellerophonts, do not appear in the record until Trempealeauian time, the latest division of the Late Cambrian. There they are represented by genera that are pleurotomarians in the broadest sense such as Dirhachopea Ulrich and Bridge, Taeniospira Ulrich and Bridge, and Sinuopea Ulrich, and by two peculiarly sinis- tral non-pleurotomarian genera, Scaevogyra Whitfield and Matherella Walcott, of which I shall write again in another place. The Lower and Middle Cambrian share with later beds caplike genera, such as Scenella Billings, Helcionella Grabau and Shimer, and others, which are thought not to have arrived at the prosobranch stage, and the anomalous group of gastropodlike forms, Pelagiella Matthew and its allies, that may not be gastropods at all. Although all specimens of Strepsodiscus major seem to show some asymmetry and always in the same sense, the species appears to be quite variable in the degree that it is developed. The holotype probably shows more pronounced asymmetry than any other specimen in the collection. Many smaller specimens are very nearly sym- metrical. The other characters are so very much those of a bellero- phont and more particularly of the Cyrtolitidae that the slight lateral asymmetry must be accepted, as it is in two or three other bellerophont. genera. However, the asymmetry in the sinistral sense occurring so early in the record may be highly significant. There is a possibility that Strepsodiscus may prove to be congeneric with Protoscaevogyra Kobayashi (1939, p. 286), in which case the NO. 3 BELLEROPHONT GASTROPODS—-KNIGHT 3 name would be submerged in synonymy. However, the genotype of Protoscaevogyra is known from a single specimen so poorly preserved that it is impossible to discover significant characters. It seems likely that Protoscaevogyra was applied to a sinistral example of some species of Pelagiella. Sinistrality seems to be a common variation in some species of that genus. I place it, along with Proeccyliopterus Kobayashi and Parapelagiella Kobayashi, as a subjective junior syn- onym of Pelagiella. It is my opinion that Kobayashi did not have opportunities to investigate thoroughly the shell morphology of Pelagiella and that his supposed genera are not distinct from it. Both of the two genera erected here have a superficial resemblance to Pelagiella. In fact, some unknown worker had labelled the types of Chalarostrepsis praecursor in the National Museum as “Pelagiella.” But although Pelagiella seems to have a shallow and broad peripheral emargination, it does not have a deep, V-shaped sinus as in Strepso- discus nor a deep slit as in Chalarostrepsis. Furthermore, Pelagiella has only two whorls or less that expand more rapidly than in those of either. It is, of course, quite asymmetrical and, in the types of the genotype species at least, in a dextral sense. Suborder BELLEROPHONTACEA Ulrich and Scofield, 1897 Family CYRTOLITIDAE Ulrich and Scofield, 1897 Genus STREPSODISCUS Knight, new genus Genotype, Strepsodiscus major Knight, new species. Diagnosis —Discuslike bellerophontiform gastropods with a deep, V-shaped sinus in the anterior lip culminating sharply at the sharp dorsal crest but without a slit. The final whorl is out of contact with the coil and, at least in the only known species, there is a slight, varying degree of asymmetry in the sinistral sense. I am placing Strepsodiscus in the Cyrtolitidae because of the sinus culminating at a dorsal carina or “notch keel.” The light contacts of the whorls and the rather steep umbilical slopes speak likewise for that assignment. That it is a bellerophont in spite of its slightly sinistral coiling is attested by the essential bilateral symmetry of its coiling, by its anterior V-shaped sinus, and by the traillike posterior margin of the aperture. STREPSODISCUS MAJOR Knight, new species PLATE I, FIGURES Ia-m Description—A moderately large species, all observed specimens of which are to a greater or less degree asymmetrical in a sinistral 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III sense: the umbilical slopes commonly set off from the lateral slopes by angulations and the umbilical slope on the right side being very slightly deeper than that on the left, but in some specimens, such as the holotype, the sinistral twist is greater and the circumumbilical angulation less pronounced; whorls three in number and the final 4 whorl free; surface features including probable pseudo-selenizone not well known but seemingly smooth except for lines of growth. Measurements Diameter Thickness The holotype (CUS Ne Me 114277) soe. rere 24.3 mm. 12.8 mm. A figured paratype (U.S.N.M. 114278)....... 13.7 mm. 6.4 mm. An unfigured paratype (U.S.N.M. 114281 a).. 18.5 mm. 6.1 mm. Hypodigm.—Approximately 150 specimens. Remarks.—All specimens are in a light-brown calcareous sandstone with rather coarse angular sand grains. The preservation is remark- ably good for so coarse a matrix. Many specimens are broken free. Great variation in relative thickness suggests that some specimens may be compressed laterally; however the matrix seems to be one that should resist compaction. The individual variation in other re- spects such as skewness and the angularity of the circumumbilical ridges suggests that individuals may vary in ratio of diameter to thickness too. Individuals are very abundant, as many as 15 showing on the surface of an oblong block approximately 6 cm. by 9 cm. Few individuals (only 2 out of 150) reach the size of the holotype, most of them being much smaller, narrower, and showing less skewness. Indeed, if the larger individuals did not show the skewness to so much higher degree than the more abundant smaller ones, it might have been missed. The specimens shown on plate 1 as figures Ie-g, th-j, 1k-l, illustrate smaller specimens. Occurrence.—Sawatch formation, north side of Taylor Peak, 4 miles south of Ashcraft, Colo. Collected about 100 feet above Archaean rocks in 1891 by S. Ward Loper. Associated with the numerous specimens of S. major were specimens of Kingstonia loperi Resser, Maryvillia loperi Resser (in both cases the types) and a specimen identified by Dr. Christina Lochman Balk as “cf. Louchocephalus.” All of these indicate a Dresbachian fauna, early Upper Cambrian, probably low in the Crepicephalus zone. 1 Oral communication by Dr. C. L. Balk. NO. 3 BELLEROPHONT GASTROPODS—-KNIGHT un Family BELLEROPHONTIDAE McCoy, 1851 Genus CHALAROSTREPSIS Knight, new genus Genotype, Chalarostrepsis praecursor Knight, new species. Diagnosis. —Discuslike bellerophontiform gastropods with a shallow, narrow sinus in the anterior lip culminating in a deep, narrow slit that generates a selenizone. The final whorl is out of contact with the coil. The shell is entirely symmetrical. I am placing Chalarostrepsis in the Bellerophontidae close to Tropidodiscus (=Oxydiscus Koken) which may well be derived directly from it. Tropidodiscus differs principally in that it is closely coiled throughout or even moderately involute. CHALAROSTREPSIS PRAECURSOR Knight, new species PLATE I, FIGURES 2a-c Description—A moderately large species of 3 or 4 whorls with sharply rounded but not angular dorsum, with the whorls of the coil barely in contact, and the last # whorl free. The umbilici are, of course, widely open. The slit is almost 4 whorl deep, narrow, and generates a convex selenizone bordered by faint striae. The aperture projects slightly behind in a skirtlike trail, then passes forward on the sides at an angle about 40° above the radius until it approaches to about one-sixth of the distance to the dorsum where it turns roundly to form the margins of the slit. The ornamentation consists of growth lines alone. Measurements Diameter Thickness Thesholotype CULS.N.M. 254282) 6's ee oo. i1.0mm. 4.3 mm. Awsmalle paratype (U.S.N.Me 114283))904-.5--..- - 5.5 mm. ? A large imperfect paratype (U.S.N.M. 114283).... 20 mm.* Q mm.* * With estimates to allow for imperfections, especially in thickness. Hypodigm.—Three specimens. Remarks——The only three specimens of this species known are preserved in a gray, crystalline limestone. The holotype has been largely removed from the matrix and is exquisitely preserved. It shows shape and surface features with remarkable fidelity. The slit can be seen as that adapertural portion of the dorsal crest through which the matrix within the shell was continuous before preparation, a roughened portion without a selenizone. The largest specimen is too poorly preserved for effective preparation and the smallest is too small to justify the labor. On the largest specimen the rear end of 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III the final whorl is separated something like 34 mm. from the crest of the nearest part of the previous whorl. The homologous distance on the smaller holotype is less than 1 mm. This indicates continued divergence of the final whorl. Occurrence—The three specimens were collected in 1889 by the late C. D. Walcott from boulders in the Lévis conglomerate near the cemetery at Point Lévis, Quebec (Walcott’s locality number 115a). Species recorded as being associated are Loganellus logani (Devine), Agnostus americanus Billings, Palaeostrophia elax (Clark) and Levi- sella sp. These are said to indicate the Hungaia zone of the Upper Cambrian. The Hungaia zone however has not yet been placed ac- curately in the standard Upper Cambrian sequence of the Mississippi Valley (Rasetti, 1944, p. 231). However, it is not older than the upper third of the Franconian stage or later than the Trempealeauian. In the opinion of Dr. Christina Lochman Balk? the Hungaia zone is close to the upper Dikelocephalus zone in the Trempealeauian. REFERENCES KNIGHT, J. B. 1947. Some new Cambrian bellerophont gastropods. Smithsonian Misc. Coll., vol. 106, No. 17, pp. I-11. KOBAYASHI, T. 1939. Restudy of Lorenz’ Raphistoma broggeri from Shantung with a note on Pelagiella. Jubilee publication in commemoration of Prof. H. Yabe’s 60th birthday. RASETTI, F. 1944. Upper Cambrian trilobites from the Lévis conglomerate. Journ. Paleontol., vol. 18, pp. 229-258. EXPLANATION OF PLATE 1 Ia-m. Strepsodiscus major Knight, new species. Ia-d, Im, the holotype (U.S.N.M. No. 114277). 1a, left side view. tb, posterior view to show skewness. Ic, anterior view showing V- shaped sinus. 1d, right side view, all 1. 1m, right side view, X 2. Ie-g, a paratype (U.S.N.M. No. 114278). te, posterior view, X 2. If, g, left and right side views, respectively, >< 1. th-j, a paratype (U.S.N.M. No. 114279). th, posterior view, X 2. li, j, left and right side views respectively, X 1. tk-l, a paratype (U.S.N.M. No. 114280). Left and right side views, respectively, << 1. 2a-c. Chalarostrepsis praecursor Knight, new species. The holotype (U.S.N.M. No. 114282). 2a, oblique anterior view to show slit, X 4. 2b, left side view, X 4. 2c, a portion of the carina (just to right of break shown right of center on 2b) showing selenizone with lunulae, x 8. 2 Oral communication. VOL. 111, NO. 3, PL. 1 CAMBRIAN BELLEROPHONT GASTROPODS (See explanation of plate.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 4 iyPE MATERIAL OF Wht seeCirs OF CLERID BEETLES DESs€hIBED.ByY CHARLES S€rNEr rin BY EDWARD A. CHAPIN Curator, Division of Insects U. S. National Museum (PuBLicaTIon 3977) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION APRIL 5, 1949 The Lord Baltimore Press BALTIMORE, MD., U. S. A. ee OM RAL On Ete sme ins Ol CLERID BEETLES DESCRIBED BY CHARLES SCHABEE ER By EDWARD A. CHAPIN Curator, Division of Insects, U. S. National Museum Between 1904 and 1921, while acting as curator of entomology of the Museum of the Brooklyn Institute of Arts and Sciences, Charles Schaeffer described 45 new species of beetles of the family Cleridae from America north of Mexico. Only in his 1921 paper on the genus Aulicus did he designate a single specimen as type of a species. In his early works he selected up to six specimens to stand as types of a species, and later he made no mention of the number of specimens before him upon which the description of a species was based. Occa- sionally it is possible for the reader to tell that one, or more than one, specimen was involved, from the remarks that follow the description. Mr. Schaeffer’s type material was originally deposited in four different collections: the collection of the Museum of the Brooklyn Institute (33 species) ; the Schaeffer collection, in which is incor- porated the Ottomar Dietz collection (7 species) ; the United States National Museum (4 species); and the H. F. Wickham collection (1 species). All these collections are now in the National Museum and therefore all, or almost all, the type material is there. The writer has recently studied all this material and has designated lectotypes for all species except where the original description and discussion indicate that no more than one specimen was before the describer. It has not been possible to recognize the type specimen of Hyd- nocera nunnenmachert Schf. This species was described in 1908 from a single specimen bearing the same data as that given for the type material of Hydnocera fuchsi Schf. (described on the preced- ing page of the same work). In 1917 Mr. Schaeffer suppressed the former as a synonym of the latter. No specimen bearing the name label ‘““Hydnocera nunnenmacheri” is to be found either in the col- lection of the Brooklyn Institute or in the Schaeffer collection. It is possible that upon recognizing the synonymy, Mr. Schaeffer returned the type of H. nunnenmacheri, minus its distinguishing label, to the type series of H. fuchsi. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Cir bo It will be noted further on in this paper that often the type locality of a species is given as “Huachuca Mts., Arizona,” and the data for the lectotype as “Palmerlee, Cochise Co., Arizona.” It should be explained that Mr. Schaeffer made his headquarters for the Huachuca Mountains trip at Palmerlee and that he used the more exact locality data on the specimens themselves. It might here be mentioned that twice Mr. Schaeffer introduced names into the literature through error. His Clerus bioculatus (1905, Mus. Brooklyn Inst., Sci. Bull., vol. 1, p. 154) is certainly a lapsus for Clerus bimaculatus Skinner, and his “var. rufulus” (1917, Journ. New York Ent. Soc., vol. 25, p. 131) is apparently an earlier choice of name for the variety that he described under the name of Clerus rosmarus var. virginiensis. In the following list the species are arranged alphabetically by specific name without regard to the genus in which each was de- scribed. The generic name of the original combination follows the specific name. No attempt has been made to show the present generic assignment of any of the species. antennata, Cymatodera 1908. Journ. New York Ent. Soc., vol. 16, p. 128. Type locality—Huachuca Mts., Arizona. Type series ——No indication of number of specimens. Both sexes described. Three specimens, including one male, each bearing the label type in the Brooklyn Museum collection; three additional specimens, without TyPE labels, in the Schaeffer collection. Lectotype-—The male bearing Schaeffer’s type label, with the more detailed locality data Palmerlee, Cochise Co., Ariz., Aug. 17. U.S.N.M. No. 42533. antennatus, Aulicus I921. Proc. U. S. Nat. Mus., vol. 50, p. 158. Type locality —Palm Springs, California. Type series—— ‘The male (type) is in the collection of Professor Wickham and the female (allotype) in the collection of the Brooklyn Museum.” Holotype-—The male specimen from the Wickham collection, U.S.N.M. No. 5oro1. arizonica, Cymatodera undulata var. 1908. Journ. New York Ent. Soc., vol. 16, p. 130. Type locality—Huachuca Mts., Arizona. Type series —No indication of number of specimens or sex. A NO. 4 CLERID BEETLES—CHAPIN 3 male, labeled rypE by Schaeffer, and a female, both from Palmerlee, Cochise Co., Ariz., in the Brooklyn Museum collection. A male from Huachuca Mts., Ariz., in the Schaeffer collection. Lectotype-—The male, labeled as tTypPE by Schaeffer. Its label carries the date June 24. U.S.N.M. No. 42537. arizonica, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 132. Type locality—Huachuca Mts., Arizona. Type series—“‘In the two type specimens, male and female, * * * ”? Tn addition to two specimens labeled TyPE by Schaeffer, there are 15 additional specimens in the Brooklyn Museum collection. Lectotype.—The male bearing the Type label, U.S.N.M. No. 42540. brevicollis, Cymatodera 1917. Journ. New York Ent. Soc., vol. 25, p. 130. Type locality— Arizona. Type series —“The single specimen, a female, was placed in the Dietz collection with ovipennis, which it superficially very much resembles.” Holotype-—The above-described specimen in the Schaeffer collec- tion. U.S.N.M. No. 59058. cephalica, Cymatodera 1908. Journ. New York Ent. Soc., vol. 16, p. 130. Type locality —E1 Taste and Santa Rosa, Lower California. Type series—“ * * * , collected by Mr. Gustav Beyer, to whom I am indebted for the pair.” Lectotype-——The male from the above-mentioned pair, from El Taste, in the Brooklyn Museum collection. U.S.N.M. No. 42538. dentipes, Aulicus 1921. Proc. U. S. Nat. Mus., vol. 50, p. 157. Type locality—San Diego, Texas. Type series—Type (male), allotype and paratypes designated in the original publication. Holotype-—U.S.N.M. No. 23085. femoralis, Aulicus 1917. Journ. New York Ent. Soc., vol. 25, p. 132. Type locality—Nogales, Arizona. Type series —No indication of number of specimens or sex. Two 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Waiek specimens, male and female, collected at the designated type locality by F. W. Nunnenmacher, in the Brooklyn Museum collection. Lectotype——The above-mentioned male, U.S.N.M. No. 42547. fissipes, Aulicus 1921. Proc. U. S. Nat. Mus., vol. 59, p. 155. Type locality—Tucson, Arizona. Type series—Type (male) and allotype (female) designated in the original publication. Holotype —U.S.N.M. No. 23083. flavosignata, Cymatodera 1908. Journ. New York Ent. Soc., vol. 16, p. 129. Type locality—Huachuca Mts., Arizona. Type series ——No indication of number of specimens. Both sexes described. Six specimens in the Brooklyn Museum collection and three additional in the Schaeffer collection, all labeled Palmerlee, Cochise Co., Arizona. Lectotype——A male from the series in the Brooklyn Museum col- lection, collected on Aug. 18. U.S.N.M. No. 42535. floridana, Chariessa 1917. Journ. New York Ent. Soc., vol. 25, p. 133. Type locality—Key Largo, Florida. Type series—The type specimen, a female, was collected and given me by Mr. G. Beyer.’ No mention of further material is made but Mr. Schaeffer, in 1918, presented the writer with a specimen, also a female, with identical data. Lectotype—The female in the Schaeffer collection. U.S.N.M. No. 59059. fuchsi, Cymatodera 1904. Journ. New York Ent. Soc., vol. 12, p. 216. Type locality —Texas. Type sertes—“One male in my possession kindly given to me a few years ago by Mr. Chas. Fuchs * * * .” Holotype.—A male specimen from Temas! in the Brooklyn Miusetine collection, bearing Mr. Schaeffer’s Type label. U.S.N.M. No. 42519. fuchsi, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 132. Type locality— Nogales, Arizona. Type series.—No indication of number of specimens or sex . One NO. 4 CEERID BEETLES—CHAPIN 5 specimen, apparently a male, labeled TYPE in the Brooklyn Museum collection; six specimens, males and females, in the Schaeffer collection. Lectotype-—The specimen bearing the TyPE label, in the Brooklyn Museum collection. U.S.N.M. No. 42539. furcatus, Colyphus 1904. Journ. New York Ent. Soc., vol. 12, p. 218. Type locality—Brownsville, Texas (San Tomas, Esperanza Ranch). Type series ——‘Four specimens in collection of the Museum of the Brooklyn Institute of Arts and Sciences.’ Four specimens labeled TYPE and eight others in the Brooklyn Museum collection and three in the Schaeffer collection. Lectotype—A male, one of the four bearing TyPE label, mentioned above. U.S.N.M. No. 42521. granulatipenne, Enaplium (sic!) 1904. Journ. New York Ent. Soc., vol. 12, p. 220. Type locality—Brownsville, Texas (Los Boragos). Type series —“Types, six specimens in the Museum of the Brook- lyn Institute.’ As there are in all eight specimens from Los Boragos in that collection, it is not possible to distinguish the six referred to in the original publication. Schaeffer had placed type labels on two specimens, a male and a female. Lectotype-—The male, selected by Schaeffer, mentioned above. U.S.N.M. No. 42525. humeralis, Clerus 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p. 155. Type locality —Tulare Co., California. Type series—“ * * * , two specimens in the collection Dietz.” One specimen, apparently a female, labeled Type by Schaeffer, in the Brooklyn Museum collection; another specimen, also apparently female, in the Schaeffer collection. Lectotype.—The specimen labeled as TyPE by Schaeffer, U.S.N.M. No. 42531. laevicollis, Cymatodera 1908. Journ. New York Ent. Soc., vol. 16, p. 130. Type locality—Huachuca Mts., Arizona. Type series —“One female.” A female specimen, without locality 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III label, is in the Brooklyn Museum collection. It bears the following labels: (1) TYPE; (2) Catal. No. 563; (3) Cymatodera laevicollis Schaef. Holotype-——The above-described female specimen. U.S.N.M. No. 425306. latefascia, Cymatodera 1904. Journ. New York Ent. Soc., vol. 12, p. 216. Type locality—Fort Grant, Arizona; New Mexico. Type series —‘‘Three specimens; a female in the [U. S. National] Museum collection from the first named locality, kindly given by Mr. Schwarz, another female from New Mexico and a male from Arizona without definite locality. Type No. 8163, U. S. National Museum.” Holotype-——tThe specimen from Fort Grant, Ariz. U.S.N.M. No. 8163. maculicolle, Pelonium 1904. Journ. New York Ent. Soc., vol. 12, p. 219. Type locality—Brownsville, Texas. Type series—‘‘Four specimens in the Museum of the Brooklyn Institute.’ Three males and one female, each labeled TYPE, in the Brooklyn Museum collection. Lectotype—A male from the above-mentioned series. U.S.N.M. No. 42524. mesosternalis, Lebasiella 1908. Journ. New York Ent. Soc., vol. 16, p. 135. L ype locality—Huachuca Mts., Arizona. Type series—No indication of ame of specimens or sex. One specimen, labeled Type by Schaeffer, in the Brooklyn Museum col- lection ; a second, without type label, in the Schaeffer collection. Lectotype-—The specimen, apparently a female, in the Brooklyn collection. U.S.N.M. No. 42543. nigrescens, Enoplium 1904. Journ. New York Ent. Soc., vol. 12, p. 221. Type locality—Brownsville, Texas (Esperanza Ranch). Type series—“One specimen in the Museum of the Brooklyn Institute.” Holotype-—The above-mentioned specimen, a male, U.S.N.M. No. 42526. N NO. 4 CLERID BEETLES—CHAPIN nigrescens, Hydnocera 1909. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 15, p. 381. Type locality—Southern Pines, North Carolina. Type sertes—No indication of number of specimen or sex. One specimen in Brooklyn Museum collection collected May 21, 1907, and two specimens in the Schaeffer collection taken May 30, 1912, and May 17, 1915, all from Southern Pines by A. H. Manee. Lectotype.—| Probably holotype], the specimen, a female, collected May 21, 1907, in the Brooklyn Museum collection. U.S.N.M. No. 42544. nigrina, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 134. Type locality—Huachuca Mts., Arizona. Type series —No indication of number of specimens or sex. A single specimen in the Brooklyn Museum collection from Carr’s Peak, Huachuca Mts., Ariz., July 20, and labeled type by Schaeffer. Lectotype | Probably holotype], the above-described specimen, apparently a female. U.S.N.M. No. 42542. nigriventris, Aulicus 1o2T. roc, Ul Ss. Nat) Mus, vol. So) p.) 156: Type locality.—Mexico. Type series—Type (male), allotype (female) and paratypes, designated in the original publication. Holotype —U.S.N.M. No. 23084. niveifascia, Hydnocera 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p. 156. Type locality—Palmerlee, Cochise Co., Arizona. Type series ——No indication of number of specimens or sex. Eight specimens, all from the designated type locality, in the Brooklyn Museum collection. One specimen was labeled TyPE by Schaeffer. Lectotype-—The specimen labeled type by Schaeffer, apparently a male, U.S.N.M. No. 42532. nunnenmacheri, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 133. Type locality—Nogales, Arizona. Type series—“ * * * , one male kindly given me by Mr. F. W. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. im: Nunnenmacher.” No specimen bearing this name was found either in the Brooklyn Museum collection or in the Schaeffer collection. Holotype.—Lost. See introductory remarks for discussion. obliquefasciata, Cymatodera 1904. Journ. New York Ent. Soc., vol. 12, p. 215. Type locality—Brownsville, Texas (Esperanza Ranch), from Acacia flexicaulis. Type series——“Three specimens in the Museum of the Brooklyn Institute.” Four specimens in the Brooklyn Museum collection, three of which, a male and two females, bear Schaeffer’s TyPE label. Lectotype-—The male, from the above-mentioned series. U.S.N.M. No. 42517. pallida, Cymatodera 1908. Journ. New York Ent. Soc., vol. 16, p. 128. Type locality—Huachuca Mts., Arizona. Type series——No indication of number of specimens. Both sexes described. Four specimens, two males, one female, and one of un- certain sex, in the Brooklyn Museum collection; one female in the Schaeffer collection. Lectotype-—A male in the Brooklyn Museum collection. U.S.N.M. No. 42518. pallipes, Macrotelus terminatus var. 1908. Journ. New York Ent. Soc., vol. 16, p. 128. Type locality—Brownsville, Texas. Type series—‘The moderately large series which I have taken or raised from branches of Acacia farnesiana and Acacia flexicaulis * * * ” Four specimens, two males and two females, in the Brook- lyn Museum collection and four in the Schaeffer collection. Lectotype——A male from the series in the Brooklyn Museum col-— lection, U.S.N.M. No. 42534. palmii, Clerus 1904. Journ. New York Ent. Soc., vol. 12, p. 218. Type locality—Senator, Arizona. Type series—“One specimen kindly given to me some years ago by Mr: Chas; Palm *°' *7-” Holotype.—The specimen in the Brooklyn Museum collection from the designated type locality. U.S.N.M. No. 42522. NO. 4 CLERID BEETLES—CHAPIN 9g parviceps, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 134. Type locality—Senator, Arizona. Type series—‘One female given me some years ago by Mr. Charles Palm.” One specimen in the Schaeffer collection. Holotype-—The above-mentioned female specimen. U.S.N.M. No. 59066. peninsularis, Cregya 1917. Journ. New York Ent. Soc., vol. 25, p. 132. Type locality—Santa Rosa, Lower California. Type series—No indication of number of specimens or sex. One female in the Brooklyn Museum collection and a pair, male and female, in the Schaeffer collection, all from Santa Rosa. Lectotype-——The male specimen from the Schaeffer collection. U.S.N.M. No. 42548. peninsularis, Cymatodera to04. Journ. New York Ent. Soc., vol. 12, p. 214. Type locality—San Felipe, Lower California. Type series—‘‘Two specimens from Mr. G. Beyer in the Museum of the Brooklyn Institute.” Both specimens are male. Lectotype.—One of the two above-mentioned specimens. U.S.N.M. No. 42516. pinus, Clerus 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p. 155. Type locality—‘“Carr’s Peak, Huachuca Mountains, Arizona, beaten from pines at an elevation of 9,000 feet.” Type series—No indication of number of specimens or sex. Two specimens from Carr’s Peak, one bearing Schaeffer’s TyPE label, in the Brooklyn Museum collection. Lectotype——A male, bearing Schaeffer’s type label. U.S.N.M. No. 42530. pusilla, Hydnocera 1909. Mus. Brooklyn Inst., Sci. Bull. vol. 1, No. 15, p. 381. Type locality—Nogales, Arizona. Type serics——No indication of number of specimens or sex. Three specimens, each labeled TyPE by Schaeffer, in the Brooklyn Museum collection. Three, without such labels, in the Schaeffer collection. IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. it Lectotype-—A male, one of the three specimens in the Brooklyn Museum collection. U.S.N.M. No. 42545. quercus, Clerus 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p. 154. Type locality—Palmerlee, Cochise Co., Arizona, from oak. Type series——No indication of number of specimens or sex. A single specimen, a male, from the designated type locality, bearing Schaeffer’s TYPE label, in the Brooklyn Museum collection. Lectotype.—| Probably holotype], the above-mentioned specimen. U.S.N.M. No. 42529. ruficollis, Monophylla 191i. Journ. New York Ent. Soc., vol. 19, p. 121. Type locality—Arizona. Type series—“ * * * , the single specimen, a female in my col- lectioner ce cae: Holoty aii female from ESBEOIE, in the Schaeffer collection. U.S.N.M. No. 59060. santarosae, Cymatodera 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p Lon On to Type locality—Santa Rosa, Lower California. Type series.—“ * * * , male and female, collected by Mr. Gustav Beyer, to whom I am indebted for the pair.” A pair in the Brooklyn Museum collection, which were labeled type ¢ and tyPE 9 by Schaeffer. Lectotype-——The male from the above-mentioned pair. U.S.N.M. No. 42528. . simulans, Hydnocera 1908. Journ. New York Ent. Soc., vol. 16, p. 133) Type locality—Huachuca Mts., Arizona. l'ype series—No indication of number of specimens or sex. One specimen labeled Type by Schaeffer, in the Brooklyn Museum col- lection, together with three others, all from the designated type locality. Lectotype—The specimen labeled tyPE by Schaeffer. U.S.N.M. No. 42541. INO? A! CLERID BEETLES—CHAPIN 1k subcostatus, Clerus thoracicus var. 1917. Journ. New York Ent. Soc., vol. 25, p. 131. Type locality—Enterprise, Florida. Type series —“Enterprise, Florida (O. Dietz). I have another specimen from New Braunfels, Texas, which agrees with the Florida specimen in the characters given above.” Lectotype-—The Enterprise, Fla., specimen, in the Schaeffer Col- lection. U.S.N.M. No. 59061. tricolor, Hydnocera 1904. Journ. New York Ent. Soc., vol. 12, p. 210. Type locality—Brownsville, Texas. Type series—‘Four specimens in the Museum of the Brooklyn Institute.’ Of these, one is a male, the other three are females. Lectotype—The male from the above-mentioned series. U.S.N.M. No. 42523. tristis, Hydnocera 1909. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 15, p. 381. Type locality—Huachuca Mts., Arizona. Type series-——No indication of number of specimens or sex. A single specimen, a male, labeled Type by Schaeffer, in the Brooklyn Museum collection. A second specimen in Schaeffer collection. Lectotype-——The male in the Brooklyn Museum collection. U.S.N.M. No. 42546. uniformis, Cymatodera 1905. Mus. Brooklyn Inst., Sci. Bull., vol. 1, No. 7, p. 151. Type locality—Prescott, Arizona. Type series —“‘ * * * , one male, which I owe to the kindness of Mr. Charles Fuchs, of San Francisco, Cal.’ Holotype.—The above-mentioned specimen, in the Brooklyn Mu- seum collection. U.S.N.M. No. 42527. van dykei, Cymatodera 1904. Journ. New York Ent. Soc., vol. 12, p. 217. Type locality—California (Los Angeles Co.). Type series—“Two specimens kindly given to me by Dr. Van Dyke to whom this species is dedicated.” A male and a female, each labeled type by Schaeffer and both in the Brooklyn Museum col- lection. I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE. Ssik = Lectotype-—-The male specimen mentioned above. U.S.N.M. No. 42520. virginiensis, Clerus rosmarus var. 1917. Journ. New York Ent. Soc., vol. 25, p. 131. Type locality.—Virginia. Type series ——No indication of number of specimens or sex. A single specimen in the Schaeffer collection. Lectotype.—The above-mentioned specimen. U.S.N.M. No. 59062. vn Wy ee ooh OY) ui le oS eV 4) ore SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 5 Roebling Fund 1948-1949 REPORT ON THE 27.0074-DAY CYCLE IN WASHINGTON PRECIPITATION BY €.G. ABBOT Research Associate, Smithsonian Institution (PusticaTion 3980) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION MARCH 8, 1949 The Lord Baltimore Dress BALTIMORE, MD., U. S.A. Roebling Fund 1948-1949 REPORT ON THE 27.0074-DAY CYCLE IN WASHINGTON PRECIPITATION By G. G: ABBOT Research Associate, Smithsonian Institution In Smithsonian Miscellaneous Collections* I have set forth an apparent cycle of precipitation at Washington associated with the sun’s rotation. In 1948, for the fifteenth consecutive year, the aver- age precipitation for predicted favorable days has exceeded the aver- age precipitation on all other days of the year. The results for 1948 precipitation are given in table 1. TABLE 1.—Statistics of Washington precipitation, 1948 Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Year Average ¢ Pfd. .. 0.292 0.049 0.274 0.146 0.386 0.123 0.095 0.173 0.086 0.022 0.190 0.221 0.175 per day \ Other . 0.038 0.091 0.087 0.007 0.161 0.140 0.136 0.338 0.151 0.165 0.195 0.084 0.137 Rati eyetessins 982016 7.68 0.54 3-15 20.8 2.40 0.88 0.70 0.51 0.60 0.13 0.97 2.63 1.28 Motall “ppt. is.e 1.616 4.99 2.05 5.31 2.44 8.59 3.98 3.60 8.00 3.63 3.11 5.78 4.93 56.41 INioguell ASsenoeac 3-55 3-37 3-75 3-27 3-70 4.13 4.71 4.01 3.24 2.84 2.37 3.32 42.26 PEKCentiz.is..ss:0 006 141 61 142 Fics 96 76 200 12 10 244 153 133 Lines 1 and 2 give the average precipitation in inches per day for preferred and all other days of the year 1948. Line 3 gives the ratio: “preferred” J ines 4 and 5 give the total precipitation and normal other precipitation in inches, and line 6 gives the percentage of normal pre- cipitation falling in the several months, and in the entire year. Preferred days had a higher average precipitation than other days in the 5 months January, March, April, May, and December, and in the year as a whole. The other 7 months went the other way, but November had almost an exact balance between preferred and other days. However, if on June 19, August 1, September 21, October 5, and November 28 precipitation had been delayed a few hours and occurred on the next following days, these 5 months would also have been favorable to the preferred days. Statistically, the ratio: prefers is expected to be 1.42. The other ratio for the 15 years ending with 1948 has actually been 1.46. Table 2 gives the dates for 1949 when the average daily precipi- tation is expected to exceed the average daily precipitation on all other days of the year. In the first column are given in Roman numerals the day number of the 27 days of the cycle when higher precipitation is expected. The remainder of the table gives the actual dates in the different months which correspond to these Roman 1 Smithsonian Misc. Coll., vol. 104, Nos. 3 and 5, 1044; vol. 110, No. 4, 1948. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 5 I 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III numerals, in other words the “preferred days” for 1949. These preferred days should give, on the average, higher precipitation per day than all the other days of the year, but there is no high probability that precipitation will actually fall on any individual day among “preferred days,” or will not fall on any individual other day. Tape 2—Predicted dates when average daily precipitation should exceed average daily precipitation for all other dates in Washington, D. C., for the year 1949 “Preferred” cycle places Jan. Feb. Mar. Apr. May June Tg AA oikvsporetts ce 24 20 19 15 12 8 i bel ag Se ee 25 2 20 16 13 9 1G Rs eee 26 22 21 17, 14 10 Wier eis cies wrens 27 23 22 18 15 II Meee ota I, 28 24 23 19 16 12 ONT itehe score alenets 8 4 3 26 23 19 BOTT seysrevsterseces 9 5 4, 31 2 24 20 DSW Sie eee II of 6 2, 29 26 22 OOF alia cesses vader seers 13 9 8 4 1, 28 24 NWA: Ge reteyees 14 10 9 5 2, 29 25 EXONC IO beycus, « attoyes 18 14 13 9 6 2, 29 DOOD eae eae 22 18 17, 13 10 6 RENEWED ers cya setahevere 2 19 18 14 II Gi “Preferred” cycle places July Aug. Sept. Oct. Nov. Dec. | oh eerste 5 1, 28 24 2I 17 14 1 BG eee 6 2, 29 25 22 18 15 10 ee 7 3, 30 26 23 19 16 QV eee 8 A, 31 27 24 20 17 ‘Wie eae rin 9 5 I, 28 25 21 18 DOIN are evs eeiateie 160 12 8 5 1, 28 25 ONT ees aes 17 13 9 6 2, 29 26 OVE er Sera OF 19 15 II 8 4 1, 28 XLT crags, Cavereteves 21 17 13 10 6 By30 EV TU dere ete cheusie 22 18 14 II 7 4, 31 D5, 40) ee Peas 20 22 18 15 II 8 DONO co sonic 3, 30 26 22 19 15 12 Dee le Se 4, 31 27 23 20 16 13 The statistical tabulation of 27-day cycles, on which table 2 is based, began January 1, 1924, and it indicated that the true cycle is 27.0074 days. In 25 years since December 31, 1923, there were 9,132 days. In 339 cycles of 27.0074 days there are 9,155.50 days, or in round numbers 23 days more than in the years 1924-1948. Hence the first day of the 340th cycle begins on January 24, 1949, as given in table 2, corresponding to Roman numeral I. It should be emphasized that this prediction relates only to Wash- ington, D. C.? 5 This paper was finished on January 18, 1949, immediately after receipt of Weather Bureau data for December 1948. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 6 Roebling Fund A PREDICTION OF WASHINGTON TEMPERATURE 1948 (MADE JANUARY 1948) BY C.G, ABBOT Research Associate, Smithsonian Institution (PuBiicaTIon 3982) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION MARCH 8, 1949 The Lord Baltimore Press BALTIMORE, MD., U. 8. A. Roebling Fund A PREDICTION OF WASHINGTON TEMPERA- TURE 1948 (MADE JANUARY 1948) By C. G. ABBOT Research Associate, Smithsonian Institution ABSTRACT The author in January 1048 predicted 55 dates in 1948 as near dates of minima in Washington temperatures. He predicted that between successive pairs of these dates higher temperatures would occur. He predicted that the average excess of these higher temperatures over those of the predicted dates would be 7°1 F. On January 109, 1949, the Director of the Astrophysical Ob- servatory checked the prediction and found that in 48 of the 55 cases higher temperatures did intervene, and that their average excess was 6°06 F. The author shows by a graph that the main part of the brief fluctuations in tempera- ture ordinarily ascribed to “weather” is caused by a periodicity of 6.6456 days in solar radiation, rather than by terrestrial complexities, as generally sup- posed. The author claims that similar successful predictions may be made for any station and, if desired, for many years in advance. The predicted dates of minima are approximate, and may be out of phase by 1, 2, or rarely 3 days. If predictions are limited to I or 2 months in advance, however, displacements of phase may be nearly eliminated. A prediction of approximate dates for Washington minimum temperatures for 1949 is given. In January 1948 I made a prediction relating to temperatures at Washington for the entire year 1948. I then sealed the prediction and placed it in the safe in the Smithsonian accounting office, with the provision that it was to be opened on or about January 20, 1949, by the Director of the Smithsonian Astrophysical Observatory. This paper contains the text of my prediction for 1948, the veri- fication of the prediction by the Director of the Astrophysical Ob- servatory, comments on the outcome, and my prediction for 1949. PREDICTION FOR 1948 In Smithsonian Miscellaneous Collections, vol. 107, No. 4, 1947, I disclosed an average period of 6.6456 days in temperatures of Washington and other cities, identical in length with a periodic vari- ation of solar radiation. Due, as I suppose, to complex terrestrial influences, the phases of this period shift somewhat as to tempera- SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 6 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III ture, though the phases do not shift in solar variation. However, it seemed to me probable that the period could be used for long-range predictions of temperature, though such predicitions may sometimes be out of phase by from 1 to 3 days. I propose to make a predic- tion of Washington temperature for the year 1948 now [January 22, 1948], seal and deposit it with the Smithsonian Treasurer, to be opened and checked by the Director of the Smithsonian Astrophysi- cal Observatory in January 1949. The enclosed graph? contains plots of the departures from normal temperature at Washington for the years 1945-46-47. From a marked repetition of the 6.6456-day period of March 1946 I observed the phases of minimum temperature at that? time. Assuming no changes of phase, I indicated corresponding dates of minimum tem- perature for all repetitions of the period from January 1945 to December 1947. As these indicators in most cases bounded curves of higher temperature between each pair of indicators, I was en- couraged to continue with my project. Using the method described on page 4 of the paper above cited, and allowing for the fact that now I am concerned, not with “zero” dates but with dates of minima, I compute the following dates in the year 1948 as likely to be at or near dates of minima in the march of temperature at Washington: ON) MER Soap Gime Ti, BLY air July.))....« 25, 40), 4aSr B22mee LBs] eaten ir Oi ace 5) 7 a oe Aug......: 4 2 1S eae Marys... ee Ol eros, 524. 7-31 Sept...... 7% 13 20\) B2omee Apr pee, el guy 20 «27-0 Fe Octs i..: 3/10) “16> S2gteee May. cite Su tG, size 23 120 Nov. 3.3.55) “22. SOUS Jeney a5 85. 02. SOL 25 4 Dees os5062 ® | 15. -22 Ribs To verify the prediction I am about to make, the departures from normal temperatures at Washington, as published monthly by the United States Weather Bureau, should be taken for each of the dates tabulated above, and also the departure occurring at the highest temperature which occurred between each pair of dates. Then the mean value for the year 1948 of the departures for the dates above tabulated should be found to indicate a lower mean of departure temperatures than the mean departure of the said dates of “highest temperature.” I will also predict what the difference will be in degrees Fahren- 1 Not here reproduced. 2 See figure herewith. NO. 6 PREDICTION OF WASHINGTON TEMPERATURE 1948—ABBOT 3 heit between the two means of departures just indicated. For this prediction I take values from the three years 1945-46-47, selected as specified in the next preceding paragraph. The computed differences of these means are as follows: 1945, 696; 1946, 8°6; 1947, 6°90; Mean, 7-1 I predict the “Mean” just stated, for 1948. Should the method of prediction succeed in Washington, it will probably succeed equally well in all other places. If any should be bold enough to use it, and wish to estimate how much difference in temperature between predicted maxima and minima is to be expected, he should consult figure 5 of the publication above cited and note that the range between maxima and minima varies several fold in Washington for the various months of the year. VERIFICATION OF THE PREDICTION January 19, 1949. This is to certify that I have caused examination to be made of the sealed package deposited by Dr. C. G. Abbot in the Smithsonian safe in January 1948. I have tabulated the temperature departures from normal at Washington on the 55 dates specified therein, and the temperature departures at the warmest dates intervening between the 55 pairs, all as published in Weather Bureau Form 1030. I find that on 48 occasions warmer dates occurred between the dates speci- fied, and that for the entire series of 55 dates the mean excess found thus was 6°96 F., as compared with 7°1 F. predicted by Dr. Abbot. ky Bs AEpRiGEH, Director, Astrophysical Observatory. COMMENTS The test proposed in the preceding pages came out nearly as ex- pected. But I realized soon after the manuscript was deposited, in January 1948, that the test might perhaps give an exaggerated im- pression of the amplitude of the 6.6456-day temperature fluctuation. For by using the highest temperatures which intervened between the predicted dates of minimum temperatures at Washington, there can be no question that some of these higher temperatures are made higher than they otherwise would be by local terrestrial meteorologi- cal influences, not directly caused by the solar periodic variation. Such extraneous influences may, it is true, have correspondingly 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III raised the temperatures at the predicted low dates. If so the error would be compensated. But one is inclined to think that this would not always be the case, and that the higher intervening values are higher in the mean than they should be in comparison with the values occurring on the predicted dates. I sought a method to estimate the average increases of tempera- ture between the predicted dates which would give too low a mean value of the excess. For this purpose I plotted the daily departures from the normal Washington temperatures from January 1 to April I, 1946, as shown in the accompanying figure. Then with a scale divided into intervals of 6.6456 days, I marked, as shown by short, heavy vertical lines in the figure, the dates which seemed best to represent the minima of the 6.6456-day periodicity at Washington.’ Inspection indicated that January 17.0000 would be a proper date for the zero phase. From this date I calculated all corresponding phases from January 1, 1940, to December 31, 1948. Assuming these 495 dates to be exact dates of minimum temperatures at Wash- ington, | computed yearly tables of temperature departures as ar- ranged in groups averaging 6.6456 days in length. In these yearly tables of 55 lines each, approximately two out of every three lines were 7 days long, and the others 6. To arrange the short lines sym- metrically I wrote 45 lines with the first space unfilled, and the rest with the seventh space unfilled. Table 1 gives the mean yearly marches of the 6.6456-day periodicity TasLe 1.—Mcan march of 6.6456-day temperature groups at Washington in the years 1940-1948 Mean temperature marches Year TGQAG 8s. ees —o?02 .—o°1r —o?8 —r1°05 —o°20 +1209 +0234 lk chee 2.63 253 1.76 3.65 BEBE 3.18 Te Tae te? 0.22 2.16 2.98 2.06 3-49 2027 2.29 Boe rahoene 1.84 2.62 2.78 2.80 mastes 1.00 0.76 ares 0.61 1.49 2.69 2.38 2.85 2cEh 1.02 ee a 1.86 1.84 2.09 BG) 2.51 2.40 2.24 Oi ne eee 0.72 2.16 4.62 4.31 3.36 2.78 2.64 ihe AP or —0.69 2.89 25i% 2.00 2.47 2.38 1.03 Toss sei. zt 1.49 2.25 1.93 1.45 Ths 3-44 3.02 Wisehal pases 1.08 2.20 2.62 AASG/it 2.46 2.59 1.99 8 It used to be the view of most meteorologists, and perhaps still is, that the brief fluctuations of temperature which we associate with the term “weather” are caused by terrestrial complexities. I think the figure shows plainly that they are mainly of solar origin. ZA Ee ue PRESB Reese s = ES cal ee V\ : a bial aa a Ee hi ae del eokales areas trsey Fa pee |e PO a BN Lay 5 7 8 10 12 1 Jha SY 4 16 [8 short, thick, indicated by |_| BEE) iS Se ile: POPE CCS ia i iia mw Pr. ib fo ie 4 fa iN Ns | al Bee yA / h | | Se BE ae eae 2 4 2 4 26 2 Fic, 1—The 6.6456-day periodicity in Washington, D. C., temperature. r cur a epa Ss al te r ya et I - Burez ‘or i <6-day period is indicated by short, thick, vertical lines. ye _ 7c ai epar res é -) “ A els I¢ (6) see W eather Bu eau it orm 1030). a he 6.6456 day Pp 10C€ A Lowe urve, the d ily departures from normal temperature, Jan. 1 to Ap 940 ( iniere nd the mean of them for 9 years. Upper curves, average temperature departures for 55 repetitions of the periodicity each year, i : ’ | ‘ ) ; ' : ; } : - gee Crees are A < iy } , ae ee ea ) i j A sat | ! 1h eh | [*y t ce oT Oe a 1 ly Sl ry Y's ty } 7 \y fy | ye (A at eee en | \ | a es Pe | Sy | } ; yt i | +} | \, | Ky St Re i a 2 } } se ‘ \ sot Tr ' / AE Tee TE ee a b Ny { H 1 \ | ; } | { — i 1 oe 2 r Se eS 1 eee BS Poti rN SLoQ iad Lasers driv, oar cgaly he PANTIES 153 att aK: of aleraqeinss sy ee : : NO. 6 PREDICTION OF WASHINGTON TEMPERATURE 1948—ABBOT 5 in Washington temperature departures for the 9 years 1940 to 1948. The upper part of the figure shows these results and their mean graphically. It is satisfactory to see that the mean curve is nearly smooth, with the correct phases for maximum and minima. The range of the mean curve is 1°63. This is certainly too low a value to be a correct estimate of the amplitude of the periodicity. For, as shown in my paper cited above, the phases of the periodicity are subject to a sort of backlash effect, and they are apt to occur I, 2, or rarely 3 days too early or too late. Hence where, as in my tabula- tion just referred to, no allowance is made for these shifts of phase, the real mean amplitude of the periodicity must be much greater than computed. I have computed also the average range per year after the method set forth in the test paper above. I give these results for the Q years in table 2. They are very closely all the same, and give a mean of TABLE 2.—Average amplitudes of the 6.6456-day temperature variations in Washington, 1940-1948 BVGaATS a eleven 1940 I94I1 1942 1943 1944 109045 1046 10947 1048 Mean Amplitudes ... 5°81 6295 6°55 6°56 6°58 6°60 8°60 6°00 6°96 6°73 6°73. For the reason mentioned above, I consider this value too high an estimate of the amplitude of the 6.6456-day periodicity in Washington temperatures. But I do not think it as much too high as the other one, 1°63, is too low. However their mean, 4°18 F., may be nearly of the right order of magnitude. While yearly forecasts of dates of minimum and maximum tem- peratures by this method may be out of phase by from I to 3 days, they may be made with equal success for many years in advance, as my earlier paper which covers 35 years shows. If the forecaster should content himself, however, with forecasts only a month or two in advance, and should take into account the phase prevailing when the forecast was made, he might hope to be within 1 day of correct phases for minima and maxima. See, for instance, in the lower part of the figure, how closely the phases held true from February 5 to March 31, 1946.4 The method, as I have said, is probably applicable everywhere. 4 Note also by the curves of the upper part of the figure that in the years 1940, 1941, 1945, and 1948 the phases were prevailingly appearing 2 days or more late. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. (iit PREDICTION OF APPROXIMATE DATES FOR MINIMUM TEMPERATURES IN WASHINGTON, D. C., FOR THE YEAR 1949 Following the method described in this paper, I give the follow- ing dates expected to be close to dates of minimum temperatures in Washington in the year 1949. Toso. ‘Jan, se. a MEE SS 2409) 31 Jeiky cao 2 0, 16°" 230 en HebwnssealO otSa B2OL B20" Gee Aug: So T2aee Tea oe Miateatten 5, ST CmmET OMS Sue nene Sept. 22.7 14.260, S270 Aor. casa 7. AeA 20) 327 Oct. »: 0.4 5 TO. e17) Seen May. 4) “1017 24 530 Nov. 5.10) 13) 110) 26eee June 240° “1g? ag), 260°". Declan 3s 9. "16" S22 1950, Jan. 2.1.45 11 (Shes The departures of temperature from the normal, as published by the U. S. Weather Bureau, Form 1030, may be taken for these dates, and also for the warmest days that occur between them. Then the average excess of these warmer temperatures over the mean of the temperatures of two cool dates preceding and following each warm one will be found about 6°9 F. for the 55 cases of the year 1949. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 7 Roebling Fund MONTEZUMA SOLAR-CONSTANT VAbUES AND THEIR PERIODIC SOLAR VARIATIONS BY Cy GZ ABBOE Research Associate, Smithsonian Institution WE-INCp _0e* ®es (Pustication 3981) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION ARIE 19 1949 The Lord Baltimore Press BALTIMORE, MD., U. 8. A. Roebling Fund MONTEZUMA SOLAR-CONSTANT VALUES AND THEIR PERIODIC SOLAR VARIATIONS By GC. GVABEOT. Research Associate, Smithsonian Institution We are convinced that solar-constant values from the Mount Montezuma, Chile, station are more accurate than those of any other Smithsonian station. This results from the meteorological superiority of the location. In three recent papers* (treating respectively of the 6.6456-day period in the solar radiation and in weather, of the trig- ger action of depressions of solar radiation to set off West Indian hurricanes, and of the effect of ionic bombardment of the earth to diminish solar radiation received here at times of great sunspot ac- tivity) I used the daily solar-constant values of Montezuma exclu- sively. The inclusion with them of less accurate data from our other stations would have been injurious in these studies of very small solar changes. In volumes 5 and 6 of Annals of the Smithsonian Astrophysical Observatory, and in my paper “A Revised Analysis of Solar Con- stant Values” ? the 10-day and monthly mean solar-constant values from several Smithsonian stations were combined in researches on long periods in solar variation. It seemed advisable to me to make a new search for long solar periodicities, using Montezuma data alone. I wished especially to test my former conclusion that all the periodic variations are integral submultiples of 273 months. I have prepared a table of 10-day and monthly mean solar-constant values for Montezuma alone, from September 1923 to December 1947. They are given in table I. In table 1 the year and month are given in column 1. In column 2 appear the 10-day and monthly mean values of the solar constant, from Montezuma observations alone. Column 3 gives the number of days entering into these mean values. Readers should note that values in column 2 are to be understood as prefixed by the figures 1.9 1 Smithsonian Misc. Coll., vol. 107, No. 4, 1947; vol. 110, Nos. 1 and 6, 1948. 2 Smithsonian Misc. Coll., vol. 107, No. 10, 1947. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 7 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. TIt Tasie 1—Ten-day and monthly means, Montezuma solar-constant values Values given assumed to be prefixed by 1.9. Thus, 1.9536, etc. 1923 192 1926 1927 1928 Sil sgh 0 2 1 Goh: 5) iisek pSg)er 6:1 479.8) | oe Tr s3%° 2S II 470 1 1G uid) 12 Il 427 46 TL» 442" wa III 540 8 TI e718: 95 Tih 303) 38 PT 3S saG Vi arom va! M 536 25 M 508 11 ME 4351-19 M 447 23 M 426 8 I 6 a I i 6 I 10 446099 |x 442. 5 4 317 10 7 445 10 452. 9 1 “rom +7 Ti apo: 2 II 396 10 II 440 8 ID, 487,26 PEL saz 4) 76 III 360 I Itt 398 0 PEE as iux Tit i300h me 7, M 436 22 M 444 8 M 369 29 M 446 25 M 442 22 mre .302" 5 2I1I— o Sele god) 8 Sraly earn esto Dr) ol Acre Uieast 7, 1h Cty 2 Li 4s een a: IT tar2h a8, Ti assures, III 490 4 Tlie S73 3 III 407 7 III 442 11 III 461 8 443 16 584 5 M 4o1r 19 M 424 29 M 448 17 Tame AZO 13 3 SS 7s Goel ses nud 9.21 (4t4)) 68)" 12, soommea Piedas™ 6 II 492 6 ie 432 e eG II 428 10 II 446 7 III 29: 8 Ti SaZzees INGO aRGy 5 III 471 7 LT | A0sueees M 368 17 M. 524 17 M 430 18 M 436 25 M 465 14 1924 1929 I 416 7 yO Geter ks 7 Sle A30) 10 1o I 481 o re ers) I II 443 8 WUE ehy | £3} ies no I 5438" 95 LDS syeyss = 2} Tit 459 9 THE 4777) 3 rh 9453 rr UD Be ine G5 TLE 48sen M 441 24 M 524 19 M 432 30 M 443 21 M 532: 46 2ielen 300) eS fH 7S Sol 472 0s me Ik ZI). 3 2. Jl, 4432) II 460 I Ee 7484) 20 We FigeY s) Ti 446" 5 Ii” 3590) 5 TL 422 6 III 479.68 III 475 6 TT) 4367) 35 TE 2336 M 396 15 M 476 25 M 466 17 M 445 18 M 385 1r S523) 20 GoaE, 1420)" 4 9) SL 44ts SO) etzele 47 0n 7 3) Laas T3804 7, Tie 496) 25 306.05 Mil yeyeey 7 7A It 3607) 9 Ui Aga: 7 TII 480 5 TII 457 8 TED Shire 9 ih aces M 453 23 M 469 14 Mi 437) 322 uM 421 18 M 393 17 192 Ak 302) 45 AL spi@y 3 TOM 373) 50 Te 4425 6 4.1, 37040 IT 416 «= Li sro} 6 it 423° 70 Tiles 43740 5} II 492 9 PTAs 2) eo III 436-8 ih 374558 TUE Var ons IIL 442, 20 M 417 21 473° 716 M 392 27 M 431 16 ME A352 rele AG? 180) Sel 4770 1 J 350) 45 2 I 480 6 Sel aa 2G 6 II 493 7 Il 442) 46 EL357; 0 UT, 433) ae, TT) V4gomes III 506 ro III 431 8 III 405 2 III 470 I III 436 10 M 489 26 M 452 23 M 364 16 456 14 M 430 21 SCOPES S 4 ae sO mils 2A Su aT zee +3391) 03 Biplues OA 7, 61305 ae TE 492" 5 II 470 10 i359) 18 Mite iG BR Th “344 ares TTI s22 6 DT 473) 78 Tit 370 I III 468 6 TOMI eee So M 526 18 M 487 26 M353) 912 M 466 18 M 388 18 1927 vf We Sie Ge IL | Mise FS T ly 396) 50 Qerleraso us 7 30 7a II 544 10 tl, esoo, ‘6 WE SBF TT (4a (us II 407 9 III 470 9 Li ass" V9 INO Rs ce) 9F} TIT aos ao III 420 9 510 26 M 467 23 M 385 14 M 417 22 M 409 26 Sle a2) Bes Die ul» 42000 a6 2 yas) va BU SAS 6 8 sla soomure Lares II 487 10 Ch 467, 7 IN owe TE 368 ee Tin 39085 Wl 470 6 III — o III 468 9 III 402 6 M 442 18 M 464 22 M 424 11 470 26 M 399 20 9 I 462 5 x2 T 482 10 a —— ro Grell eaz2eneo 9 Ur 30700ne ASS) ey, II 460 3 II 500 6 int eg AA Tess Sco III 431 8 Ill 497 7 tle 466) ao Th AGor ees lib 4r6eas M 457 20 : 484 20 M 479 15 M 471 16 M 397 24 192 Foe A536. 415 A730 ee 400 44 Guns 7,1, 446) 10) to sto satay My e524 18 Il 499 7 II 472 9 Tt acne Ir 4aomara III 528 x1 III 390 5 Mite Zi 8 Wa eee Tt 3700s 528 24 M 461 19 M 445 22 M 434 22 M 412 14 Tiel A567, 25 405) ‘2 Ball rete 3 8) 4a 5 tr I 4x90 II 494 8 II 41o 5 Ti y426, 955 Diy 44ore 7, Tl “4220635 III 498 6 Tile 400) wer WII 420 12 III 440 6 Ill, 480s M 520 23 408 8 423 24 M 444 18 M 442 22 NO. 7 MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT TABLe 1.—Continued 1929 1931 1932 1933 1934 12 I 468 5 3 I 390 4 6m 44659955 Oe 7473) (7, 2 477, 1S It 468 9 Tisoo88 (6 D450). 9 III 426 5 III 514 9 Tir tsos 6 TES so4s as M 461 18 M 472 22 M 486 17 M 470 21 1930 1935 I 460 2 Annie s20 5 7 lee g2 as TO 1) 403) 93 I Meare, 1-4 T2462) ta DP 3000 LES soo) ) 7, TII 395 2 VUE Ey) G III 435 10 Tl 4900))47, M 420 8 M 472 14 462 16 M 408 17 2 I 398 6 5 I — o 8) iey4s6) 7. tr I 1438) 45 2 II 460 I II 520 I II 412 5 II 509 8 Mit 74800 35 III 460 2 TR 38251 as ARNE ay os i 437, x2 M 480 3 42I 17 M 484 14 Byla'sOny 17, Cpela4 73 e7, 9 2159 432 5 2 isira os 3 TM) (422) 8 II 480 2 II 447 6 II 480 4 III 443 10 Tile t4s2 9s MURE Msey AG III 505 10 438 25 M 459 14 M 445 16 M 502 19 1934 4 I 444 8 7 7 OV 2 ee LOE para 5 I 504 5 4 ihe t4or 7 Ti sosae 3 IE 310) | 6 Ine” Agcy) 71 Th 443° 6 TIT sos) 194 Tee 3468 5s IU Gags M 430 21 M 549 9 354 16 M 486 11 5 I 450 4 Soe 4850 ate 707, 2 alesSS5in 2 5 liies49g0) ) 4 Tisr266 Tit 34250 S 1h ABR 1B WIT, "470. 4 Tileras6s a7 ihe s3775 06 ThAGZun 16 IIl M 470 12 M 482 17 M 361 21 M 477 10 Gunlg4637 63 Qunlems020 Oe t2cl 983738083 3) 20430) 73 6 Fe 2462, 4 Ties Sie is Tass eara TMepzcsen Ss III 483 3 ess 7) 27 es Ses TUES 436 M 469 10 M 520 18 M 411 10 M 496 14 1933 Vlas7a 68) TO) is445 47 I 510 I 4 1ss03)) 33 7 II 490 1 Ty iso. | 6 ia 7 2s 15 Tieeas6n Ss Ps.) vs Tieeas2 | 38 PiTgeAsse 2 Ihiova3s= 38 M 462 12 M 482 21 M 480 8 M 455 19 Speleea7sauesy lt. iwasra. 7 2a 7S 5 I 480 6 8 i. 490.) 4 ieE4s2) 0S Tio 48o "4 TES 4660) 5 TT: 479) Ir 460) 57, III — o III 440 11 481 18 448 22 M 479 7 M 457 22 9 I 456 10 12 I 444 8 3) lerason 4 6 “he s18) 1G 9 ie z400) 2 TS 477/01 67 T3731) 3 Te 1605 5 wile, 422. 4 III — o III 407 4 DitaGounes M 433 16 459 15 M 410 10 M 508 13 1932 Torelert53. Oo Tele aG2en a! 4 I 410 8 > ees eae E7, 10 Migeazor | <6 iehaa7s 3S Mea4s571 9 Tisss5o25016 III 462 8 ETES46s5) 2 Tie370) 10 Wt) 477. 57, M 461 23 458 9 Miia tr 27 M 501 20 We ole4s2: 5 2) lia ss) 4 hE 8) OL ssoyee 4 II Dias. | S TI 492 4 1 4or 9 II 497 6 DE 2528) 5 IGE 7igys TUL ary 5 III 499 7 M 492 18 M 453 12 M 397 24 M 500 17 Tae lencson 0 sules63 rs 6 Sissazt 18 9 I 474 10 12 Tipsss5. 10 II 447 7 OTs Weep 3 II 490 9 Tis esko. 4 Ti Saag, 107 This yat2) \6 Thi SA660 0.7, III M 540 23 M 431 17 M 420 25 477. 26 1931 I 430 1 4 197407) 4 7 leeatSe dy to lew4gs nS I niee4o 740 6 Dig 4415 9 19; DATS DUSPAO TES 7, Iil — o IQUE iG) eR} TL 449 © © 7. gHHO ras ef) M 484 7 M 455 16 451 19 M 503 23 PY IU vty, 3 Ge vig 8 I — o iT Tevssoe | O 2 II 407 6 Tie arn 4 5 Lipe42874 6 i Gea oR III 458 5 III — o TLE S420) 19 Titessogeess 480 13 Mira zon o Mangesears M 520 22 cl ” ONTO NN HR MNSPNO Ads on cel al bv co OMmo iS} mood ” Lol be AOon WHwWO oano b co) CN COW Wh COW Lal La ORR DA WOuwk cooc#h Noun ol Lal KH “ bw COA RHA OMOHN HOAHR LOWS ~ co bd woo rd ONNOQ Cm COM bd RHO OOM ” RONDO AcnhoO MN bd bd wNnn Oo ro] DAN w SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. bit TABLE 1.—Continued I 1938 1939 1941 a 472 8 Quelen462)) SO taser e375 3 575 4 Tiras T se a7, II 448 8 TLS 4 7/3 eZ, II 545 4 TT a3 33 III 479 10 III 445 9 WOE BRGY Go M 460 18 M 464 24 M 435 22 M 549 17 1940 7 I — o 10 “1 4701'.'S Tel aAGom as 4c agsooao TT vAc6gus II 495 6 Las Om ans II 520 4 DTT yak Oa a7 Ril 522 5a10 IMUM + AIRS ye sabe INE Gre M 457 15 M 497 23 M 454 19 509 15 8 Tisor 9) tr SE i4osis 6 2 ras 7 ay, Bs Wess omuey Taz 200 II 532 10 II 434 10 II — oo III 484 7 TSS 7) ees III 385 4 LLEe6o02 eno M 483 25 M 521 19 M 426 21 S75 13 ule aoui7y) , 22el Misra) us 3 _I 469 8 Go Gee GS II 487 9 Wk eth Gy) II 362 10 TES 575 0a III 404 10 III 493 10 III gor 7 IV Gara A M 445 26 M 514 24 407 25 MY 9s75.0x 1939 1o I 419 8 1 I — o 4) Laas) ty x2 7) WE O37 es Ll asa 0 II — o IT 458 4.0 Lias6o geno FIT 5401045 III 400 2 LOTS 44 ei WOE Re 8 M4523 M 400 2 M 489 18 M 572 17 Or IL CORP Pyle ee MEST Ol IG: 8 1 s667 7 IP 4as7 iG II — 0 TL 3486 728 Tieusazanr LIED ssosien 6 III 463 6 IMHE BiG 66) III 480 6 M 490 21 442 13 M 506 26 M 529 17 Tape eso es G) de ZE) ) 6 I 486 10 9: Litsgon V8 TE W570) 193 II 442 5 WE igi LE S00 m7 Wil 54x 8 TA 2 MOL BCG) 9S MOE CR 8 es 540 16 443 16 496 24 532 17 1938 Teel eAo00N 1 ApethadA27 ans GD AL Bae 10) Le sass TL S534 0s Tesi ms: T4953 96 479 7 Tiles 20 ees TL rg'27 es: IME aR G6) TS sins 7 Beep sie M 437 18 M 514 21 506 22 2 I— oo Hb AeRyY 6 Sia Ts SoSh 6 pee GS TT 4400 92 IAG etsy 7G) II 484 8 D472 III — oo TUL so3 eens TNS TofS TIE soon 440 2 M 397 23 M 496 19 M 496 21 eg} de Bye Grblso3cy OL 527 76 12) 525 as Oy ay 3 Ih sete {3 T5647 Tl 402s med THOM Vieyey 58 Ti ee40 20009 WUE Zz), PIs "548 eeno M 461 8 391 24 528 21 529 21 t042 41 445 8 7. 420° 38 LOMAS OMe TS Grass Il 432 6 Ths 388" io LN LGR BAG) Es 9 III 462 6 TL 305i III 486 7 TIT 49400 es M 446 20 M 402 23 476 13 M 537 18 BG, EARS) Selets733 <8 Tr oLesdos 6 2 oLiag6u a7 BM aie) 5 T3540005 Me Zee) TES 400 ny III 306 7 TG ee ts TUE Aras: TLE 460 sano 430 21 M 385 a1 M 426 22 M 486 20 Gili4azo) 6 Onvlena7Gur7, T2) er4zouey, 3) ean ees ieasaeees T4477 a6 II 492 5 II 409 10 III 460 4 TD yac2 es IMS ey7 vA Pilea aens M 442 15 M 459 21 M 468 16 M 419 18 1941 Veal e4ar 8 ‘10 Tiga Deas 3 7 4 I 424 § LE Age) 8 It 42 7 TES33anes II 464 10 III 437 6 Tiles 200no III 410 I Til + 426\an7, M 441 22 M 399 23 M 504 16 M 443 22 820) 462 5 17s bes Someia 2 .leis2s era eel ieey II 466 8 It “z¥sye). II — 0 Dl “4909 7. III 458 6 TIT 404 8 Tite sgoueG III 482 9 403 19 WE Shp ke M 564 10 M, 473 20 NO. 7 483 475 464 474 494 490 464 482 450 461 436 449 451 443 450 449 443 440 456 447 439 485 526 483 429 404 447 428 404 419 390 406 465 486 443 467 425 501 439 461 427 443 459 441 463 467 465 465 503 497 408 500 484 443 406 476 on ON RH is) b&b b ONN ~ Oo nonwm © 0Wo NOW oO TABLE 1.—Continued NH OnNHA HHMN CO | al hr OW nohH Co 1S) CO C0 OQOHA CORN ROWH bd 468 432 364 420 455 436 470 453 422 410 446 435 420 413 440 417 503 435 480 477 451 458 415 448 451 488 487 475 469 453 481 467 466 436 418 442 - ty 1S) Lad PhP OHM NOOK ONS orn ~ iN} en) mUanoo mnacnaed & WwW oano ~ bo BOBWOW ONAD Oars is) be kROoOONM MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT 1947 421 8 2 409 7 II 391 8 III 407 23 M 375 10 pee 410 7 II 489 7 III 418 24 M 378 «65 4 I 360 3 Il 357m iil 365 15 M 3857 Bik Rik en) II 393 7 III 378 22 M 492 9 GRE 488 5 II 370 «5 Til 459 19 M 486 5 FR al 486 8 II 433) 6 Til 469 19 M 432 4 8 I 470 5 II 458 5 III 455 14 M 484 5 © i 495 6 II 406 8 III 455 10 M 409 8 1o I 396 7 II 420 2 Til 405 17 M 444) 9 rt Y 450 6 II 415 4 III 439 17 M 388 = 5 2) 421 7 II 428 7 TII 415 19 M 437. 4 395 4 270 «+I 400 9 goo. | = — o Spi 9S 37300 ao 492 4 450 1 Azan 480 8 360 410 420 409 342 382 413 381 380 450 458 439 450 443 4Il 430 384 470 454 435 428 378 424 414 373 387 416 393 422 432 395 418 429 433 479 445 479 454 438 460 433 454 435 442 Ow wm AAW - ~ bo Led ounmno Do CO ANN DN I Lad WbuUun nuns i) bo whup Ob OM WNINO i] 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III to give the complete solar constant in calories per square centimeter per minute. Figure 1 shows graphically in curve A the march of the monthly mean values given in table 1. Curve B, on the same scale, gives departures from 1.945 calories remaining after 14 periodicities speci- fied in table 2, below, have been removed from the original data given in column 2, table I. Table 2 also gives the yearly mean values, and numbers of days entering into them. It gives also smoothed-curve values derived from these yearly data, after plotting them as shown in figure 2. In the statistical search for periodic variations reported below, the smoothed- curve yearly mean values of table 2 were first to be removed by sub- traction from the original monthly means. In order to do this the smoothed yearly means were first expanded graphically into a plot of smoothed monthly means. I do not take space to publish these smoothed monthly means, as their simple derivation will be easily understood, and as it makes no appreciable errors in the periodicities, to be given in table 2, whether these smoothed monthly means for eliminating yearly changes of the solar constant are the best that could be found or not; for these periodicities are found as means from statistical tables including many repetitions of the periods, and local errors are smoothed out. In previous analysis of solar-constant values * numerous periodici- ties in solar variation were found to proceed simultaneously, all being approximately integral submultiples of 273 months in length. I did not wish to adopt this master period of 273 months in this present research without independently confirming it from Montezuma data alone. Figure 2, however, itself seems to indicate that a period of about this length would fit the yearly variations of the solar constant. There are researches of other authors which support the validity of a period approximating two II-year sunspot cycles, as being in evi- dence in various solar and terrestrial phenomena. Thus G. E. Hale discovered that magnetism in sunspots reverses its polarity in a re- markable way with each successive sunspot cycle of 11 years, so that the sun’s magnetic condition is restored only after two 11-year cycles pass, or about 22% years. A. E. Douglass has remarked a 23-year period in tree-ring widths. Various meteorologists have found it in terrestrial data. I myself pointed out that Wild’s meteorological studies of the Russian Empire, when supplemented by later data, showed very clearly a 23-year cycle in weather at St. Petersburg. * Ann. Astrophys. Obs., vol. 6, p. 181, 1942; Smithsonian Misc. Coll., vol. 107, Ne. 10, 1947. j cosmo PE ee | UA | UA TW ie SAL | i / = tt St 6 a aT Coase mime I it is a | | [Pe ae aia ete eles le YE VV ae Fic. 1.—A, monthly mean march of solar constant, December 1923-December 1947, Montezuma station; B, residuals after periodic fluctuations removed. PT Ea COR CEOCe SEIoG ba iaiall CIMA I Se CONT NAIL is MI — NOE 7, MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT 7 Nevertheless, I began this present research without assuming a 273-month master period. First of all I removed the yearly varia- tion from the values in column 2, table 1, as noted above. I then plotted the residual values and found that by far the most prominent periodic variation displayed in a large-scale plot of the residuals was of about 39 months. Seeking to fix its length as accurately as pos- sible, by careful inspection of the large-scale plot, I finally decided on 394 months. I am not sure that the period may not be 39 months, which is exactly 1/7 of 273 months; for the presence in the data of many other periodicities, and of accidental errors of observation, makes fixing of the exact length of a long period doubtful. Never- theless, a table was prepared of seven columns, alternately of 39 and of 40 months in length. The mean of these columns is plotted in figure 3, c. As the reader will see, the march of this 395-month peri- odicity is nearly a regular sine curve, and its amplitude is 0.0069 calorie, more than one-third of 1 percent of the solar constant. The 393-month periodicity was removed by subtraction to give a second list of monthly residuals. These also were plotted on a very large scale. There showed then a periodicity of considerable ampli- tude, approximately 91 months in length. A table 91 months long of three columns was made from the second residuals. With so few columns entering into the mean it seemed best to smooth the mean values by 5-month running means of them. The smoothed values being plotted, the 91-month periodicity appeared plainly, but super- posed thereon there appeared a period of § of 91 months. As it would be preferable to determine this curve of about 15 months by itself at a later stage, a smooth curve was drawn of g1-months period, cutting symmetrically through the 15-month superposed excrescences. The g1-month periodicity had the amplitude 0.0054 calorie. It is not of sine form, but rises rapidly to maximum, and falls slowly to minimum, like the well-known sunspot frequency curve of II years. This 91-month periodicity was removed from the data, leaving a third list of residuals, which were plotted on a large scale. The third list, when plotted, showed clearly a strong periodic fluc- tuation of about 68 months. This was determined by forming a table of four columns, taking their mean, smoothing it by 5-month run- ning means, and plotting the smoothed means in a curve given in figure 3, b. Very clearly there is a period of 1/7 of 68 months super- posed on the principal curve. Not wishing to evaluate a 9}-month periodicity until a later stage, I drew a smoothed curve as shown in figure 3, b. It is nearly of sine form, and has an amplitude of 0.0053 calorie, slightly under one-third of 1 percent of the solar constant. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIL oO It was now apparent from the behavior of the yearly variation of the solar constant, the excellence of the 393-, 9I1-, and 68-month periodic curves, and the superposition of curves of 91/6 and 68/7 months, as noted above, that it is quite justified to regard 273 months as a master cycle in solar variation, and that many periodicities, nearly or exactly integral submultiples of 273 months, exist simul- taneously therein. In all my subsequent search for periodicities in solar variation, as displayed in Montezuma solar-constant values, I accepted the 273-month master period, and sought for integral sub- multiples of it. 45 | 42 Lb 1941 1924 26 26 30 32 34 36 38 40 42 +4 46 (948 Fic. 2—Yearly march of solar constant, 1924-1947. Proceeding by the methods explained above, the periodicity of 544 months was next sought, found, and determined. Its amplitude is 0.0020 calorie, its form, like that of 91 months, comprises a rapid rise and slow fall. The curve, though smoothed by 5-month running means, has excrescences indicating the encroachment of a period ap- proximating 8 months. Study of it was postponed, like those found with the g1- and 68-month periodicities, for later determination. Attempts were then made to determine periodicities of 454, 34, and 3043 months. But these proved so far dominated and obscured by variations of shorter periods that they were all passed over for the time. However the curve drawn when seeking a periodicity of 303 months clearly indicated a periodicity of half that length, of fairly large amplitude. So the next search made concerned 15% months. It will be noted that solar variations of 273, 91, 68, and 543 months period had now been extracted from the monthly data, and that the fourth list of residuals was now being used. NOT MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT 9 A period of 15$ months is 1/18 of 273 months. It was now practicable to divide the data into three groups, and tabulate them in 6-line tables of 15 columns.* In this way it could be decided if the supposed 154-month period continued in all three sections of the interval of 273 months. Figure 3,a, gives the mean curves for the three tabulations and the general mean. The three group means Saas roe fated Poeelo ea Zaneel eZ 42 pad AB | polled _CEea eee Ra? eet : eb Ee 4 Che ee ee ee ee yes A ceeeeu 2eeeeeeae er eS es Fic. 3.—Examples of solar periodic fluctuations. a, 154 months. Observations 1924-31 ; 1932-39; 1940-47; and 1924-47; b, 68 months. 1924-47; c, 394 months. 1924-47. show no certain secular displacement of maxima and minima, have nearly similar forms, and nearly equal amplitudes. Hence their mean was taken as shown in figure 3, a, and is regarded as a very well- determined periodicity of solar variation with an amplitude of 0.0030 calorie. This mean curve, being well supported in detail by the group means, is used unsmoothed, and the departures of it from 4 Whenever a periodicity not of exact months is determined, values or columns are omitted occasionally in tabulations, so that the mean values of columns fit the exact length of the periodicity. Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.) TIT 1.945 calories were subtracted from the fourth list, giving a fifth list of residuals. Though convinced of the validity of the assumption of a 273- month master cycle, I have passed over any discussion of the sun- spot cycle of 114 years, approximating one-half of the master period. I now take up its consideration before noting the discovery of sev- eral other periodicities. Figure 2, which displays the variation of the yearly means of Montezuma solar-constant values, does, indeed, show depressions at the years 1925, 1931-32, 1937-38, and 1944. These may indicate a sunspot-cycle influence, but might better be attributed to the 68-month cycle which has already been discussed. Moreover, these depressions appearing in figure 2 are very small, with ampli- tudes only about 1/12 of 1 percent of the solar constant, yet the 68- month curve, when specifically determined as given above, has an amplitude approaching 3 of I percent. Meteorologists recognize that the 11-year sunspot cycle is reflected in temperature, precipitation, and barometric pressure. Aldrich, also, has shown ® by the study of individual daily values of the sunspot numbers, and of solar-constant values, that there is a complex cor- relation between these phenomena. But my residual plots of monthly solar-constant values do not show any 136-month periodicity of ap- preciable amplitude. This is not really in contradiction to the find- ings of meteorologists. It is well known that the sunspot areas bom- bard the earth with electric ions. These, by acting as centers of condensation for water vapor and dust in the earth’s atmosphere, may very well be competent to produce meteorological changes. Be- sides this, the ozone contents of the atmosphere may be affected by them in a way to influence meteorological phenomena. So we may recognize two kinds of solar influences on meteorology. One de- pends on variations of the solar radiation, the other on variations of ionic bombardment. Having discovered and evaluated periodicities of 273, 91, 68, 544, 394, and 152 months in the variation of solar radiation, as evidenced by monthly mean solar-constant values of Montezuma, I next used the original 10-day mean values to seek for periodicities of less than 12 months. For such short periods the longer ones hitherto discussed produce no sensible interference. It would be tedious to recite all these trials. The method was always the same. By means of a long paper scale divided at regular intervals to represent a suspected period, I tested on the long plot of 10-day means whether such a ® Smithsonian Misc. Coll., vol. 104, No. 12, 1945. NO? 7 MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT ie period seemed to be likely. If it seemed so, I arranged the 10-day mean values in groups of tables, each comprising about one-fourth of the total interval 1924-1947. They were never less than six lines long, and with as many columns as there were 10-day intervals in the proposed period. Where periods were not exact multiples of 10 days, values were omitted, or columns were omitted, occasionally, to bring the average lengths of the lines to that of the proposed period. The criterion of a true period was always that the several group tables agreed substantially in their means, as to phases and ampli- tudes of the suggested period, throughout the whole 273 months. Such good agreement is shown for the 154-month period in figure 3, a. In several cases proposed periods failed to meet this test, and were rejected. Sometimes the phases shifted regularly from group to group through the 273-month interval. In such cases the period was shortened or lengthened to give unchanging phases. As a result of this branch of the investigation, periodicities of 5 2/15, 8.035, 93, 114, 11 15/16 months were recognized as true, ac- cording to the above criterion. Being incommensurable in length, there was no need to subtract them one by one from the data. They could not materially influence each other. After determining them in the 10-day mean data, they were transformed into monthly means. Then their marches were tabulated throughout the 273 months, their amplitudes added algebraically at each month, and the algebraic total per month was subtracted from the fifth residual list, remaining after removing the longer periodicities named above. This left a sixth list of monthly residuals for further exploration. To shorten a tedious story, the methods explained above, when applied to the sixth list of residuals, discovered additional periodici- ties of 144, 194, and 244 months. When all had been removed from the data, no other periodicities seemed worth investigation in the residual plot remaining. It is plotted as curve B of figure 1.6 The mean of the departures from 1.945 calories in curve B is 0.00189 calorie, or 0.097 percent of the solar constant. Many of the larger departures, which materially raise the mean as just given, occur in 6 One disturbing feature will be noted in figure 1,B. Though the year 1947 shows no remarkable eccentricity in curve A, it gives a great slump of 4 per- cent in curve B. This is strange, for all the periodicities seem to fit the last year’s data, including 10946, as well as the earlier years, as we see from figure 1,B. One notes, however, that curve A of figure 1 is almost entirely below 1.945 calories in 1947. It may be that the Montezuma values of 1947 are subject to a yet undiscovered error. Further observations of future years will decide. Ln Lom) Lo | VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS N Loz 00z boz 61z Sbz zgz “4y6r = G b £ zb6r 261 Sbz lzz 60z gzz biz 961 fiz goz gL SLI gor gzz £12 Stz gz giz 6bz I op6r 6 8 4 9 $s v £ z I of6r 6 8 Z 9 $ bz61 zIi- s/s OW ——s FS — I— ¢€+ S+ 9+ 9+ 9+ 9+ $+ & + 1— $— 6— orm G— L£— wzI— LI— 1z— ra lz— fe) 4— bi— ¥ g— $+ g+ or z— Oo gy -- b+ to) b— “I— zI— 1+ 6— 9— o— gi—- gf— £61— ~=— =— ¢+— 9— g—) or— II-- vI— Si——s ZI————s 6I——s OZ ——ae— f&— &&— t— bte— be— be— be— fz— 7z— 1z— o— gI— 41— S1— ei— Ss zI— oI— 6— 3— 4£—- 9— s— $s — s— S$— GS$— +— F¥— F—- FH— e— €— ~—- 2e—- z— I— 0 I £ b S £ or II £1 SI gi 1z $z gz gz gz 6z of of 6z 6z ge ge le Sz fz 1z gI SI £1 It g 9 $ £ z I te) I— e— €— ze— b v £ 5 I Co} 2— = O=— F=— oO f— F1— ZI— 61— oz— 1z— &z— &ce— bz— z— y~— lz— gz— ge— gz— gz— lz— gz— Sz— tz— oz— S1— oI— 9— z— z 9 6 Z1 Pr 91 ZI gt O1 oz oz oz Iz 1z zz fz bz Sz Sz fz 4 oz “1 FI ZI or 6 8 Z 9 $ 4 z— @€—- @—- @— e— €— €— b — g— S$— S— S—_ g— 9— £-—- £6—- £4£—- 4— £— Z— L— 9— 9- s— y— e -— z— I1— ts) ° z £ Ss g OI zI fr £1 €1 €1 cI Il OI g Z 9 $ £ z I ° I— I— z— 6 9 £ I b— g— II— tiI— Ss gI— gi— oz— z~— Sz— gz— Sz— z— S1— 1i— 9— oOo £ oI Iz of gf ob Iv eb eb zP oP Ze ee ge zz SI eI II oI (@—) 4£— 4— £— oI— Sz— oO $ £ z ty) z OI ZI er £1 6 8 9 z fe— oI— s— $ o or I1— b— b— £ b— 9— z— z 1I— z fa) gI zI 9 6 9 I 4 z 6 SI $1 S g— I b— LZ— S1— b— I— S— 9 v1 g-— b £1 Si 9 (a) oI £ € oO c— ¢— g- I— z— C0) I — b £ I £— Z1— I z— ZI v 8 zi gt rai 8 £ £ £ z y— oI— z 9- z2-— 9 z— s — ¢$— o1 te) £ fe) or S £Z— z 6 — $ — I— £ S t9) I— 9— ee ee ee eS 000‘OI “9110/20 ————. JO S}IUN UI Saliojeo SP6'r wWoIZ SaInjiedap yo yo1eyVY I €z61 Joquiadeq uo psseg SUOTJVAIOSGO JUL}SUOI-IIOS vUINZdJUOP WOT Sayiipoisag pajiwjaq—z% Iv J, “*Sqo SAep JO ‘ON Ia *sqo sAep jo ‘ONT re3 x yyoous Iva tz I oF ea fz £ -S 16 v es 39 S oz FPS £ 69 FOE II gt tbe PI lz #61 gi of PST OI Sz pri fz 6z Stir te lz FII gz gr £6 be 61 S£0'3 £s It SIS €Zz Jo 00°O 8 sy}UOUI UOI}IeI} pn} ur aeu 9-day polog -txoidde jo 10}eu -1m0usq NO. 7 MONTEZUMA SOLAR-CONSTANT VALUES—ABBOT 1s the months December to February, when the atmospheric conditions at Montezuma are less favorable, and when many days are lost to observation. It cannot be claimed that the periodicities removed are perfectly correct in forms. Hence the final residuals are larger than they should be on this account also. We may conclude that of the variations of solar radiation indicated in figure 1,A, and which ex- ceed I percent in range, accidental error of observation contributes less than 2/10 percent, and the periodic variations nearly I percent of the total range. In table 2 I gave the details of the 14 periodicities in the variation of solar radiation which have been discovered. There may be others of less than 5-months period, some of minor amplitude, and still others exceeding 273 months in period, which our observations have not yet continued long enough to discover. Indeed the large fluctua- tions of Great Lakes levels occurring at intervals of about 45 and QI years seem to indicate that the double and quadruple of the master period of 273 months are of very great importance in meteorology. There is also the noted Bruckner period, of about three sunspot cycles, which may also be found eventually in solar-constant values if they continue to be observed for some years longer. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 8 SOME STAGES IN THE EVOLUTION OF tae NERVOUS SYSTEM AND tBe PORE-GUl OF Tie POLvVeriia. BY FRANK RAW University of Birmingham, England OO92 28 8oy, (Pus.icaTIon 3983) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION AUGUST 4, 1949 The Lord Baltimore Press BALTIMORE, MD., U. 8 A. ei a a SOME STAGES IN THE EVOLUTION OF Tak NERVOUS SYSTEM AND THE FORE-GUT | OF THE POLYCHAET By FRANK RAW University of Birmingham, England CONTENTS PAGE aA UTC EL GLI Street Gens ee teem: cai there ecrchehe resents isis Sig eae cig SAS Be OE Ad eet he I MIR reem Dy red OLE TUNE CON gees). aBuvo-c eu alec syeye. 6/01 dee wrara valess aculavehay enteerdiotouenn ea ols bere 2 History of opinion regarding the constitution of the brain in polychaets..... 3 Table 1. Summary of opinions regarding the significance of the brain...... 4 Significance of the divisions of the brain in primitive polychaets............ 5 IStIneSMIas he Ee UNICO ALteL VEL eld fica cic sts cosiesel ol aelsiscs A orcterd coe nei ln eve 6 sihesstomodeuml in) 2uice and other polychaets.....-s+-+2 ese oeiacile cele 9 siheystomatocastricinenvous systetm im poly chaetss 0 scmisie etree cieleielerleteierrs 10 Interrelations of the stomodeum, the visceral nervous system, and the brain AMEtMeMAN GCEStiys Ol A UsUCCme eae isles ele cies ool aici core oa 10 Table 2. Supposed stages in the evolution of the central and visceral systems @fpthe poly Ghaetec As comics gers | tans aatoctamue ser aie cents oes ae ee eens 14 Completiomvotathesbrainiy, others coca die beeieie/ a) ota..schenaione misidia eitiolate ewes Gneaete s ehaene 19 Seay Cit UNS Eb NOIES THNAAV Ga conashosdounoccd sou dpacnuasseuaoouooKd 19 Limits and significance of the prostomium of the polychaet................ 22 Morphological sienificance of the hind-braim: <7. .!-....- 0. 22 eile ae eis» re 25 Morphological significance and relationships of the peristome.............. 26 Application of the author’s theory to polychaets in general................. 27 Table 3. Comparison of supposed homologues of the morphological enti- ties of the head region of typical errant polychaets................ 30 (Corie nto’ SoA es ei ae Gian Ce aieicls los Dai atic ore een d HAGESmE GS dace 33 PRELEMEMICES RM Rise eR erect hcis ale v/s Septiie.w aise tim sinfeptin de ieee Ss pene ete > 33 INTRODUCTION As polychaets are primitive among the annelids, so are the Errantia among the polychaets; and the Eunicimorpha among the Errantia. Though the Amphinomorpha, since the work of Storch (1913), have by many been placed at the base, the Eunicimorpha are in many respects the most primitive. This is seen in the brain with its ex- tended form and marked subdivisions, and in the stomodeum with its fold form and the very small amount of its protrusibility. This paper, at its first inception in 1926 (!), was suggested by the char- acters of Eunice (Leodice) as given by Heider (1925, see p. 61). SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 8 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. iat The Amphinomorpha also are primitive; especially (as is generally supposed) in their tetraneury, in their brain (except for the hind- brain), in the absence of specialization in the buccal segments, and in the small protrusibility of the pharynx. But they seem less primi- tive than Eunicimorpha: in their brain, both by its greater concen- tration and the enormous development of the “hind” or nuchal brain; and in their stomodeum, both by its cylindrical form and greater protrusibility. The great antiquity of the Eunicimorpha is indicated by the annelid jaws (scolecodonts) which have long been known. They are abundant in Paleozoic rocks; and great numbers of species have recently been described, especially by E. R. Eller, from various horizons between the Middle Ordovician and the Upper Devonian. Practically all those described up to the present belong to the Eunicimorpha (e.g., Eller, 1945). Sometimes the denticles of both the “upper” and “lower” jaw series are found in their natural association (Lange, 1947). The claims made in this paper, however, give them a very much greater antiquity than the Ordovician. Indications of evolution are best seen in these most primitive forms; but all the errant forms furnish evidence in varying degrees. The specialized “sedentary” families are naturally less satisfactory for this study and need not be considered, but their ancestors doubt- less experienced the same evolution. For the presentation of the theories given below, it is not necessary to discuss the origin of segmentation. The evolution pictured fol- lowed the acquirement of segmentation, and points back to a ringed worm more primitive than any now existing. Moreover, it preceded the acquirement of the trochophore larval stage. The brain with its complement of sense organs and appendages, the stomodeum, and the visceral nervous system are all markedly compound in errant polychaets; and it is the chief purpose of this paper to claim that their structures are connected in origin. THE BRAIN OF EUNICE The brain by its form, especially in Eunice, suggests an origin in a complex and lengthy aggregation. Hatschek, 1891, and Racovitza, 1894, were the first to divide the brain into fore-brain, mid-brain, and hind-brain. Heider (1925, figs. 7, 12, 13, 15, 16) does the same; and he determined minutely the external features of the neuropil mass (see figs. 1, 3, 4, and 5, p. 6). These three “brains” are here claimed to constitute three distinct categories of nervous matter. no. 8 THE POLYCHAET—RAW 3 The fore-brain. This separates itself from the rest by its position, form, structure, the senses it serves, and the gap above and behind it (through which pass muscles of the mouth-lips, the so-called “palps” of many authors). It is suggested here that this was the primary brain, and as an archicerebrum may well have had a complicated history of its own. It receives the ventral root of the perioesophageal commissure; and gives origin to the two pairs of stomatogastric nerves and to nerves of the mouth-lips. The mid-brain. This, the main brain, innervates the prostomial appendages and the eyes. It is clearly subdivided into three successive sections diminishing rearwards—an anterior, middle, and posterior. The anterior receives the dorsal branch of the perioesophageal com- missure, and innervates the anterolateral antennae; the middle in- nervates the single pair of eyes and the posterolateral antennae; the posterior innervates the median antenna and the posterior eyes when present. Heider, like Pruvot, instituted only two divisions: the last two here instituted forming his posterior division; but in his figures (loc. cit., figs. 12 and 13; see also figs. 3, 4, and 5 accompanying this paper) they are very clearly distinguishable from one another. The antennae are comparable with the tentacular cirri of the second buccal segment and with the cirri of the normal body segments. The eyes too may well be homologous with the pigment spots on the body segments above the parapodia and gills. Each of these three sub- divisions of the mid-brain is below claimed to originate in a separate pair of body ganglia added to the brain at different times, and to be the only parts which have homologues in the ventral chain. The hind-brain is much smaller, having only a small fraction of the volume of even the smallest (the posterior) division of the mid- brain. It consists of two separate ganglia: these innervate the nuchal organs, which have no counterpart elsewhere in Eunice. These three “‘brains,” the fore, mid, and hind, thus constitute three distinct categories, distinct not only in character, but also, it is believed, in their origins and their histories as explained below. HISTORY OF OPINION REGARDING THE CONSTITUTION OF THE BRAIN IN POLYCHAETS As the theory here submitted differs so fundamentally from all previously presented, a short sketch of the diverse opinions regard- ing the origin and significance of the brain was written, but is with- held in order to shorten this paper. The views of the various authors beginning with Pruvot in 1885 (see list of references, p. 33) are extremely varied; they are summarized in table 1. The theory of PEL VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS ‘g10y passe1dxa SMOIA YUM Soaide ays joodset auo UT ‘quaiayIp A[PPIM a21e SuOISsN[oU0D Jaq syoadsei 19y30 ysow UT ynq ‘syUoUISIS sjusseide1 stosoqo1d 9y} 3eY} ‘ZIA _juowusas APOG Y}INOF 10 pATY? 24} 0} JOYS SBuofeq,, UINIuIOJsoId ey, “UOT[Sues eatjoouUOD |Yy} Aq peyAdouur st (slosoqoid papuazxa ey} JO JUoZas quaiedde ysiy 243) ,,[] WueurZes,, tay :Ajuo sal ayy &q poyueseidar St YoryM ,,J USUIdES,, Joy So}eAIOUUT yt pue ‘yt 0} ssuojeq wlaj}sAS SNOATIU [PIIOSIA Ol} o1seq e&,,—wajsks snoAtou jso[dulis 94} Surjuasoid se sia4anNy Spiesea1 2y4S JSOUL 94} SB UIeIq JOeYIOSI[O 94} Sexe} 94S Ja} ‘e104 pajuaseid yey} wo1Z JUeIayIp o3INb st yoo]jno sayy :yun & SI uleiq yoeqoAjod 24} Joy 0} SuIpsoo0y «PeATIep 9q UPD SUIIOF 19Y4}O 94} YOTYM Wo1y ad4} ‘a}e19uaZep oq Jsnul Pt ‘aeuUDjUe pue Safa [Je 4sO] SurAeY ‘Ysnoy, ‘!oArpwAd ‘E]L[NUUY 94} Ul WojSAS SNOAIOU 94} JO AppIqeyS OY} SI ‘SItiay “J x) puke wWeorsuLAy WoL} paAtiop ‘ajdiourid ‘epodoiyjty pue eioydoyokug ‘ezeyokjog ‘ejyeeyoost[Q ul Apog 94} JO UOoIse1 JOL1ojue ay} JO UOT}e}USUISES 94} SuIssnosip ‘porvadde aavy (gh6r ‘4P61) AUaFT "WW Piney &q sioded ‘uoreorqnd 10} payiumgqns pue paystuy sem saded sty} 20uIS , 2) ureaq 24} 04 peppy || a a o 11d Vi uleiq [eyUIUISasuOU 9Y LT, re¥ e ie) Id Wi Ila uresq dy} episinO | Uleig [eJUsUIsasuo NY urleiq [eJUIUISIsUO NY V uleig JUuSIUYy ; Joye I “daS ¢ AyUO auIos Ul poppe ‘I “sas Jaded juasa.id ‘mey O£61 ‘UOsye}sny OZOI ‘WI014s19pOS 8z61 ‘Zz61 ‘WorjsuePT QzOI ‘1ous9( pue psreulg SzO1 ‘a199WeT 1°S3S Z BIS €°SaS £°39S Z°S39S 1°33S uleIq [eJUsUISISUO NT 1 °S9S sdjed pue €°SaS Z°32S sdij yjnour yyIM I “Bas uresq yeyonu euuojue aeuusjUue aeuuajue Saqo] I141}seso}eW Jo ureq-PurH] UEIPSIN JetoqejO1I}sod Jesoze[OIOUVY -0}S 10 UIeIG-310 44 ee ee uleIg [euUDJUe IO UTeIG-PIT «upg ayy fo aauvayiubis ayy Burpépba4 suowrgo fo nun g—I AIaV [, ZIGI-1 161 ‘UOSS|INN QIO1 ‘uUsISWOF] QORI ‘ez}IAODeYy 16Q1 ‘Yayos}e yy Sggr oanig yeypsyeyy Aq pojsas -SNS }SI SUOISIAIP uleIg no. 8 THE POLYCHAET—RAW 5 the present writer, expressed cryptically on the last line, is developed below. SIGNIFICANCE OF THE DIVISIONS OF THE BRAIN IN PRIMITIVE POLYCHAETS The fore-brain. Both the morphological position and the importance assigned to the fore-brain vary greatly. By Pruvot and by Nilsson it was regarded as the first of three head segments (though the other two of these supposed segments given by the one are different from those given by the other). Hanstrom, on the other hand, is in the opposite camp, regarding it as the sole addition to the brain. To Hatschek, Racovitza, Binard and Jeener, and Soderstrom it was perhaps of little significance. But in primitive polychaets, as already remarked, it is separated from the mid-brain by an extensive muscle gap and may exhibit large size as in Eunice and Euphrosyne. In polychaets generally it gives issue to a pair of stomatogastric nerves, and in primitive forms it innervates the mouth-lips which evidently preceded antennal palps in the palpal function. This strongly suggests its great antiquity as an element of the brain. If, however, the stomatogastric lobes of the Amphinomidae and Euphrosynidae were the exact equivalent of the fore-brain of Eunice, as Gustafson seems to claim, there is a significant difference between them; for the ventral root of the perioesophageal connective enters the fore-brain in Eunice, but does not enter the stomatogastric lobes in these other families. This great discordance does not seem to have been remarked by Gustafson, and needs explanation. The equivalent in Amphinomidae of the fore- brain must also include the part which receives the ventral root of the perioesophageal connectives. The characters and relationships of the fore-brain, therefore, quite justify one in assigning it to a different category from that of the adjoining mid-brain; and the fact that in the most primitive poly- chaets it innervates the organs which function as palps, suggests that it is the primal brain. The hind-brain. The significance of the hind-brain is clarified by the discoveries of Séderstrom. It again falls in quite a different cate- gory from that of the mid-brain, as is indicated by its innervation of the nuchal organs alone. The fact, too, that in the Amphinomidae and Euphrosynidae these organs extend over several segments of the body, and that the hind-brain in them is so developed as to suggest the incorporation in it of several pairs of nuchal ganglia, suggests that the hind-brain is the latest addition to the brain. This is also 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fics. 1 to 5.—All the figures are traced from Heider’s; and the present author’s inter- pretation is added on the right side of figures 3, 4, and 5. Fic. 1.—(After Heider’s fig. 7.) Diagrammatic longitudinal vertical section through the anterior part of Eunice punctata. The oesophagus and the median plane ganglia are shown in median section; but the brain and the jaw sacs are represented as cut at one side of the median. X 12. Fic. 2—(After Heider’s fig. 10.) Oblique cross section from the third segment behind the peristome on the dorsum forward to the lower lip. X 20. Fic. 3.—(After Heider’s fig. 13.) Brain of Eunice punctata, ventral aspect. Ca. X 20. Fic. 4.—(After Heider’s fig. 15.) Dorsal view of the brain and the anterior parts of the central nervous system and visceral nervous system of Eunice punctata. Ca. X 10. Fic. 5—(After Heider’s fig. 16.) Side view of the same. Ca. & IS. no. 8 THE POLYCHAET—RAW 7 ABBREVIATIONS USED ON FIGURES 1-5 The central nervous system is in outline, the visceral nervous system is in black in figs. 4 and 5. a, nerve of the posterior single antenna. | a2, nerve of the posterior pair of antennae. | as, nerve of the anterior part of antennae. b, bearer. | bf, fore-brain. | bm, anterior section of mid-brain. | bm2, Heider’s hind section of mid-brain. | bh, hind-brain. ch, chitin thickening on the base of the prostomium. | cl, callus on the jaw- pad. | co, buccal commissure. d, dissepiment. | ec, ectoderm. | eml, epithelium of the mouth-lips. f, fang. fp, fore pad = 1 IT. gs, supra-oesophageal ganglion of visceral nervous system. | g2, infra-oesaopha- geal ganglion of visceral nervous system. | gz, hindmost oesophageal ganglion. | gc, ventral ganglion chain. | gx, x-form ganglion body. | gp, glandular pouch. jp, upper jaw pad. | //, lips of pharyngo-oesophageal rift. | / IJ, lips bound- ing the upper jaw sacs. | /JJI, lip bounding the lower jaw sacs, the lower lip. | 17, lower jaw. m, muscles of the prostomium. | mg, beginning of mid gut. | mj, muscles of the jaw sac. | ml, dorsal longitudinal muscle. | mr, ring muscle layer. | mrt, retractor of the jaw sac. n, oesophageal visceral nerve or nerves (from the fore brain). | co 1, oesophageal visceral commissure, anterior part. | co 2, ditto, posterior part; in figure 4 it is the inner pair joining in g2; the outer pair is n oe I, following. | » oe I, lateral oesophageal nerve. | 1 ph, pharyngeal visceral nerve | nu, nuchal organ. oe, oesophagus. | f: to fs, podial nerve roots. s, saw-plate. | slz, sl2, sacs of the lower jaw. | swz, su2, sacs of the “upper” jaw. | t, “tooth. The present author’s views as to the evolution of the brain and of the gut are suggested by the order of the capital letters A, B, C (respectively fore-, mid-, and hind-brain), and the Roman figures I, II, and III for the three successive invaginations and the three different ganglion pairs in the mid-brain. The original brain was the fore-brain and the corresponding gut the mid-gut. The first addition to the gut was the oesophagus—I, after which B I was added to the brain. The second addition was the upper jaw series of sacs with their armature—II, after which B II was added to the brain. The third addition was the lower jaw sacs with their armature—III, after which B III was added to the brain; which then incorporated also the hind-brain—C. The sequence of lower lips resulting from the successive invaginations are still represented by 1 I, 1 II, and | III, the present lower lip. The concomitant evolution of the visceral nervous system need not be repeated here. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III suggested by the fact that “both Kleinenberg (1886) and Meyer (1901) have shown that the ganglia of the hind-brain are marked off from the ganglia of the nuchal organ and incorporated in the brain” (Gustafson, 1930, p. 339). The same writer supposes that it will be the same in other polychaets. The mid-brain. The most primitive form of mid-brain seems to be exhibited by Eunice, where its extension in a plane and its strong suggestion of segmentation into three successive and well-defined divisions, which can be called mid-brain III, II, and I, can be regarded in that light. Hanstrém (1927, p. 592; 1928, pp. 259, 260), it is true, is very dogmatic in his expression of the opposite view; but the facts are against him. The anterior and largest pair, mid-brain ITI, innervate the anterolateral antennae (“‘palps” of authors), and receive the dorsal branch of the perioesophageal connectives. The middle pair, mid-brain II, innervate the posterolateral antennae and the eyes. The posterior and much the smallest pair, mid-brain I, innervate the median antenna, which, contrary to the view of Binard and Jeener, is best regarded as a coalescence of two; when two pairs of eyes are present, as in some eunicids, the posterior smaller pair may with confidence be assigned to this brain division. In the other primitive brains, those of Euphrosyne and of Amphi- nomidae the mid-brain is still extended in a plane, and its parts have the same topographical relations to one another, but the subdivision into three pairs is not immediately apparent. The identity, however, of the succession of nerves with similar topographical relations strongly suggests a composition of the same three divisions III, II, and I. The antennae compare so closely in Eunice with the pair of tentac- ular cirri on the second body segment and with the dorsal cirri on the other body segments, as is the case also in the Amphinomidae and in many other families, that it is quite natural to homologize them. The eyes again associated with mid-brain II and I can be regarded with great probability as serially homologous with the pig- ment spots on the body segments of Eunice. From the comparisons below instituted with the visceral nervous system, the writer believes that the three divisions of the mid-brain represent three segmental body ganglia which have been added to the fore-brain at three different times—I, the posterior, first; III, the anterior, last—and that they do not belong to adjoining segments of the ancestor. The supposed origin of these parts by the secondary subdivision of a unit brain—Hanstrém’s “fore-brain,” consisting of our mid- no. 8 THE POLYCHAET—RAW 9 and hind-brain—into four divisions, grouped above in two categories (Hanstrom, 1928, pp. 259, 260), and the grading of the three divi- sions of our mid-brain by posteriorly diminishing size is very diffi- cult to imagine; whereas on the theory detailed below it follows quite naturally. THE STOMODEUM IN EUNICE AND OTHER POLYCHAETS At the outset in this paper the Eunicimorpha have been claimed as primitive, but it may be thought by some that in respect of their stomo- deum they are less primitive than some others. Hempelmann (1934, im Kukenthal and Krumbach, Handb. Zool., vol. 2, 7, p. 161, Anne- lida Polychaeta systematic section) divides the Polychaeta Errantia into—(1) suborder Amphinomorpha with tetraneury; and (2) sub- order Nereimorpha with dineury, of which the first family described is the Aphroditidae and the last the Eunicidae. From this one might think that he regarded the Aphroditidae as the most primintive and the Eunicidae as the most specialized. When, however, we note that the systematic part is a translation of Fauvel in Fauvel’s order, except for the extraction of the Amphinomidae and its elevation into a sub- order, we may conclude perhaps that the order was without signifi- cance for Hempelmann. The stomodeum of the Eunicimorpha is unique among errant poly- chaets—in its form, its armature, and its very slight protrusibility. Its form is remarkable, consisting of the pharynx subdivided into four successive sacs separated by foldings of the pharynx wall, followed by the cylindrical “oesophagus.” All the sacs, and the oesophagus as well, extensively communicate with one another by rifts in the midline. The armature, too, is remarkable and of great variety within the suborder; but by comparisons between the various types the courses of evolution can be clearly seen. The denticles ‘are arranged within and on the ventral sides of two of the sacs in all forms, constituting respectively an upper jaw series of pairs of denticles and a lower jaw with a single (chisel and crusher) pair. The evolution of the armature could form a separate paper. In Eunice only the lower jaw chisels and the upper jaw fangs are protruded. The primitive symmetrical pairing of the denticles, well seen in Stauronereis (Staurocephalus), has been lost in the Eunicidae, s.s. In the Amphinomorpha the stomodeum, though very different, presents the nearest approach to that of the Eunicimorpha in the very small amount of its protrusion and in the fact that this is largely limited to the ventral side, recalling the condition in Eunice where Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III only the ventral lower jaw and fangs are protruded. In the Nerei- morpha (as restricted by the exclusion of these) the pharynx is cylindrical and is often far protrusible, reaching its greatest length and protrusibility in the Glyceridae and the Goniadidae, which because of their unique conical, secondarily annulated prostomium, might well constitute a suborder, the Glycerimorpha. This long, cylindrical, far protrusible proboscis the writer regards as a high specialization. THE STOMATOGASTRIC NERVOUS SYSTEM IN POLYCHAETS A visceral nervous system has been known since the time of Stannius (1831), who investigated that of Amphinome rostrata and showed that the stomatogastric nerves issued both from the brain and from the first ganglia on the perioesophageal connectives. Pruvot (1885) showed that such a system occurs generally in the polychaets, the nerves issuing in different cases either only from the brain as in Eunice, or both from the brain and from the oesophageal connective as in Nephthys and Phyllodoce, or only from the perioesophageal ring as in Ophelia. A double origin has also been shown by Rodhe (1887) for the Aphroditidae (Polynoe), viz, from the brain and from the first perioesophageal ganglion. In Nereis according to Hamaker (1898) and Holmgren (1916) the innervation is from the brain and from the suboesophageal ganglion. The visceral nervous system of Eunice, described in considerable detail by Heider, and that of Hermione described by Bernert, are faily closely comparable with one another. In all cases there are two pairs of stomatogastric nerves, one pair of which arises in the fore-brain near the midline. It therefore is probable that all are inherited from the primitive polychaet. So far as the writer is aware the origin of the visceral nervous system has not been determined or even deduced. Heider remarked upon its individuality in opposition to the remaining part of the nervous system; and he concluded that it had a separate origin. INTERRELATIONS OF THE STOMODEUM, THE VISCERAL NERVOUS SYSTEM, AND THE BRAIN IN THE ANCESTRY OF EUNICE The purpose of this paper is to show how closely related, in the writer’s view, are the stomodeum, the visceral nervous system, and the brain; and how strongly a parallel evolution of all three is sug- gested by the anatomy of Eunice. The compound stomodeal system (of oesophagus and phdryensy No. 8 THE POLYCHAET—RAW II and the visceral nervous system are closely associated with one another. Throughout their common extension the visceral nerves and ganglia are in the stomodeal hypodermis. And it is suggested that these two systems are also connected in origin; indeed, that the visceral nervous system is a direct result of the stomodeal invagina- tion; that before the invagination commenced, the present visceral nervous system would not exist, and the brain would consist only of fore-brain ! The stomodeum is very complex and the invagination that gave rise to it must have been very complex, probably much more so than we can realize (see Heider, 1925, figs. 5 to 10; also figs. 1 and 2 in this paper, p. 6). Though so complex, it divides itself quite naturally into three divisions: (1) the oesophagus; (2) the sacs associated with the upper jaw armature; and (3) the sacs associated with the lower jaw armature. Three successive invaginations or series of invaginations appear therefore to be represented, which must have been separated by long halts, and probably imply changes of feeding. The first is represented by the oesophagus; and this would seem to have very long antedated the other two, represented by the upper jaw series and the lower jaw series of pharyngeal sacs, which were separated by a shorter interval. At each of these later invaginations, the new was not merely an extension of the old, but a new structure lying ventrad of the old (fig. 1). The first or oesophageal invagination was probably a protracted process; it is represented by the long oesophagus, commencing in front of the mid-gut and bounded by the lips of the pharyngo- oesophageal rift (/ J, figs. 1 and 2), which, but without the rift, may represent the lower lip at the end of that stage. The second or upper jaw series was also probably a long process: its effects were to produce the main cavity of the pharyngeal sac with the upper jaw sac above it, and bounded below by the “fore- pad,” which again, but without its rift, may represent the lower lip of that stage. It contains the so-called “upper jaw’ apparatus, the different elements of which—the bearers, fangs, saw-plates, and “rub plates”’— are here interpreted as representing several pairs of appendages! In the more primitive eunicid, Cirrobranchia parthe- nopeia, seven such pairs of appendages might seem to be indicated ; but comparisons throughout the suborder show that it is impossible to judge of the number involved, and this is not surprising when it is realized that the evolution of the stomodeum took place very long before Cambrian time and probably over 1,000 million years ago. The third or lower jaw invagination adds a further pair of pockets I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III to the pharyngeal sacs. It contains the “lower jaw” plates, here interpreted as representing another pair of appendages added to the mouth armature; and it is bounded by a new, the present, lower lip (TIT). Each of these pharyngeal invaginations has produced a two-fold sac, each sac being partly subdivided by the pair of pads, which can be likened to parapodia, on which the pair or pairs of denticles are situated. Besides the rifts between these pads, the parting between the upper jaw sacs and the lower jaw sacs, as also that between the upper jaw sac and the oesophagus, are markedly bifid. This suggests that with the evolution of each new sac the previous lower lip became bifid owing to the necessity for through-communication through the sacs to the oesophagus. The visceral nervous system and its distribution, as suggested above, are explicable as direct results of the invaginations, which not only involved ordinary ectoderm and pairs of appendages, but also the “central” nervous system, which was still continuous with the ectoderm. It may not be possible to assign to each invagination its exact contribution to the visceral nervous system; on the other hand this is clearly subdivisible into an oesophageal and a pharyngeal section. At this early stage, too, the main nerve cords from the brain were in a more primitive position than now obtains; for, judging from the parts of the visceral system both on the oesophagus and the jaw sac, they were ventrolateral, not midventral, in position; and, that this was still the case after the evolution of the polychaet, is suggested by the far-separated ganglion chains of Serpulae and of some arthropods. Seeing that each invagination was rearward in direction, its effect on the nerve chains, when it involved them, was to pull them back into a pair of loops open in front; and their anterior connections, i.e., with the brain, were enormously stretched, in contrast with their posterior connection, i.e., with the continuing nerve cords. An effect of this is to be seen in the fact that the oesophageal visceral nerve cords from the fore-brain are quite free from the hypodermis, until, in the supra-oesophageal visceral ganglion, they reach the oesophagus. The stretch has pulled them free, as it has also the fore-brain. The open loop in the nerve cords, after each invagination that in- volved them, seems to have been closed by the advance of the fore- most of the unaffected ganglia to the brain. Such a change might be aided by the existence of a hypodermal nerve net. It is not to be expected that the whole history will now be trace- No. 8 THE POLYCHAET—RAW 13 able, for nature in such an unimaginable time would simplify a system complicated by repeated invaginations. And in any case a more detailed knowledge of the visceral nervous system is needed than is furnished by Heider, if we are to trace the history as com- pletely as is still possible. The development of these interrelations is illustrated by table 2, stages I to 10, pp. 14-15. The effect of the most ancient, the oesophageal invagination, on the disposition of the nervous system is to be seen in the oesophageal nervous system; but it is perhaps difficult to interpret. Heider de- scribed a considerable system, all of which, excepting the connection with the brain, lies in the oesophageal hypodermis (Heider, 1925, pp. 77 and 78). From the posterior lappets of the fore-brain the visceral oesophageal nerves (n, fig. 3), freed from the hypodermis, and soon apposed to one another, extend to the supra-oesophageal visceral ganglion (figs. 4 and 5, gi); thence the separated pair of nerves continue rearward (figs. 4 and 5; ” co 1), and both then branch, one branch of each ( oe J, figs. 4 and 5) continuing presumably for the length of the oesophagus in the side pads (fig. 2), the other (n co 2, figs. 4 and 5) continuing rearward to join its fellow in the infra-oesophageal visceral ganglion (g, figs. 1, 4, and 5). According to Heider the visceral oesophageal ring, thus completed, has the character of a long-stretched ganglion. From the infra-oesophageal visceral ganglion a single cord in the ventral wall of the oesophagus extends rearward to a terminal ganglion g; (at the level of the 4th podial nerves). From this a pair of nerves go to the sides of the upper jaw sac, and so join up with the pharyngeal system. The oesophageal visceral nerve ring mimics the oesophageal central nerve ring; but the union of the visceral nerves in the midventral line of the oesophagus calls for explanation, in view of the evidence for the lateral position of the nerve cords at that early period. Might it be due to mechanical drag imposed by the upper jaw sac invagina- tion? The oesophageal visceral nervous system is much more com- plex than our ideal oesophageal loop; but some redistribution may well have taken place, since the time when the oesophagus was the whole of the fore-gut. If appendages were present, and were in- vaginated, as is probable, they have disappeared, having been super- seded by those of the jaw sacs. The narrowness of the visceral oesophageal ring in contrast with the great width of the correspond- ing ring of the central nervous system can be attributed to the total absence of oesophageal appendages in contrast with the size and the action of those of the pharynx. I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES Tt TABLE 2 (with stages 1 to 10).—Supposed Stages in the evolution of the central and visceral nervous systems of the polychaet Sig a ee rs Se icy OSS ren Oesophageal Invagination ef Pi ao (Upper Jaw) Invagination a lst Pharyngeal © ST EG “ By + + open a8 “, 2nd Pharyngeal ¢% + + closed, (Lower Jaw) Pek a + + simplified —10 Invagination no. 8 THE POLYCHAET—RAW 15 Stages 1 to 10 of table 2 Stages. 1. Brain and left ventrolateral nerve cord of the annelid ancestor of the polychaet. (The nerve cord was possibly not ganglionated at this stage. The podial-longitudinal and the nuchal nerves are omitted.) 2. The oesophageal invagination has produced open nerve loops, along which lie an unknown number of ganglia. 3. These oesophageal nerve loops have been closed by the advance of their terminal ganglia to the brain to form mid-brain I 4. The oesophageal nerve loops have been simplified as suggested by the present visceral oesophageal nerves. 5. The rst pharyngeal or upper jaw invagination has produced open nerve loops, along which lie a considerable number of ganglia. 6. These upper jaw nerve loops have been closed by the advance of their terminal ganglia to the brain to form mid-brain II, which has pushed I to the rear. 7. The upper jaw nerve loops have been shortened by a coalescence with the visceral oesophageal cord. 8. The 2nd pharyngeal or lower jaw invagination has produced an open nerve loop without, however, involving any ganglia. 9. These loops have been closed by the advance of their terminal ganglia to the brain to form mid-brain III, which has pushed II and I to the rear. (In this way I has been pushed up to the Ist nuchal ganglia which will join it as hind-brain. ) 10. These nerve loops have aborted and the hind-brain is added. 10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, III The contrast between the oesophagus and the upper jaw sac could hardly be greater; and one can conclude that the interval between these invaginations was an enormous period of time. In this interval the open loops in the nerve chains would be closed; and the fore- brain would receive its most ancient addition. Judging from the structure of the brain this is now the posterior section of the mid- brain (mid-brain 1): its small size (quite out of proportion with II and IIL) suggests its great antiquity. It would carry with it, too, a pair of cirri and a pair of eyes. At this time it would lie immedi- ately behind the fore-brain (table 2, stages 3 and 4). The process just envisaged would have two advantages for the animal: first, the visceral nerves, as they had become, would thus constitute a system of their own, apart from the “central” nerve cords, according with their contrasting functions; and, second, the segmental ganglia behind the new effective mouth acquired again direct connection with the brain, thus rehabitating the “central” or rather the ventrolateral nerve cords, as they then were. The first pharyngeal invagination, comprising the compound upper jaw sac subdivided by the jaw pads, appears to be mainly responsible for the pharyngeal part of the visceral nervous system. The second pharyngeal invagination, the lower jaw sac, appears to have added only the anterior part of the pharyngeal nerves, as explained below, p. 18. The upper jaw sac, together with its bounding bifid lips above and below (/ J and / JJ, fig. 2) is highly muscular ; and, except for the muscles extending from it, it is bounded laterally by the body cavity. One of its great functions is its eversibility, allowing the fangs to be shot out to seize prey. This is largely effected by extroversion of the main long sac beneath the jaw pad. Heider was unable to trace the nervous system throughout; but he believed his detached parts to be continuous, as suggested in his figure 16 (see also fig. 5, accompanying this paper). Except for the connections with the brain, all lie in the hypodermis of the jaw sac, and are lateral in position, from which we can infer that the nerve cords were still lateral and in the hypodermis. The nerves connect- ing with the oesophageal section have already been mentioned. Sup- posed to be connected with these are the pair of main nervous centers in the pharynx—Heider’s “X-form bodies” (figs. 4 and 5; gr). In plan these are irregularly quadrangular, with concave sides and horn- like angles, and they lie over the muscular posterior bases of the jaw pads, on either side of the dividing rift and with their convexities — directed toward one another. Each consists of a nerve-fiber mass covered with a layer of small ganglion cells. no. 8 THE POLYCHAET—RAW 7 Heider, after satisfying himself as to the nervous character of the bodies, suggests that they are motor centers for the muscles of the jaw pads; and though he could not trace connections, he suggests that their posterior horns may connect with the nerves from the terminal visceral oesophageal ganglion g,, and that their anterior ventral horns connect with the two visceral pharyngeal cords, We can agree with Heider that the function of this pair of nerve centers is connected with the muscles of the jaw pads; but we can go further and suggest, on the theory here advanced, that each X- form nerve center is due to coalescence of the ganglia (lateral at that time) of the segments, the appendages of which are represented by the upper jaw apparatus, and the highly modified muscles of which are represented by the muscles of the jaw sacs. There is close coordination of the appendages, and there is union of the ganglia serving them. The two pharyngeal cords of the visceral nervous system arise from the sides of the fore-brain via the anterior roots of the oesopha- geal commissures, and extend rearward to the side wall of the pharynx. They are traced by Heider as far back as opposite the second ventral ganglion and podial nerves. Their course is along the deepest (farthest sideways) part of the main sac, ventrad of the lateral bases of the jaw pads, and closely clinging to the hypodermis. If Heider’s supposed connections are correct we have here a nerve course from the fore-brain rearward to the main visceral nerve center in the pharynx, forming one side only of a visceral loop, the other side of which is to be found only in the oesophageal section. If at one time the ancestor of Eunice had separate oesophageal and pharyngeal nerve loops, their adjoining halves forming an opposite loop have disappeared. But if connection between the outside halves was effected, the connection of the inside halves with the brain would be redundant. Such a connection might be caused by the sharp folding of the hypodermis which the stomodeum exhibits, or might arise when the lip between the two invaginations became bifid. On the general theory applied above to the oesophageal invagina- tion, the evolution of the upper jaw sac was followed by the advance of the foremost unaffected segmental pair of ganglia to join the brain, thus completing the previously open pharyngeal loop. This is repre- sented by the middle pair of the three mid-brain “ganglia,” numbered II (the second in size and the second in antiquity) of the additions to the brain. It appears to have arrived like its predecessor behind the fore-brain, and to have pushed the previous addition to the rear; for the anterior section of the “‘mid-brain,” considered below, was 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, III the last to arrive, as indicated by the fact that it bears the visceral pharyngeal nerves and the roots of the oesophageal commissures, and further, they all three diminish in size from front to rear. Each of the three parts of the mid-brain carried with it to the head a pair of cirri, and the first two also a pair of eyes; for each segment in Eunice generally bears a pigment spot with the structure of an eye just above each appendage. The cirri survive as the antennae of the prostomium, but usually in Ewntce one pair only of the eyes survives, belonging to mid-brain II. These eyes must have superseded earlier eyes innervated by the fore-brain. When a second pair of eyes exists, these are a posterior pair and belong to mid-brain I. The union of the pair of cirri belonging to this last into the median antenna may be connected, first, with the relegation of the brain division serving it to so posterior a position and, second, with its small size as com- pared with mid-brain II and III (see table 2, stages 5 to 7). The second pharyngeal or lower jaw invagination comprises the two sacs beneath the “‘fore-pads.” The upper of these sacs is bounded on its ventral side by the lower jaw pair of plates, there implanted. The median edges of these are free and thickened, forming a pair of crushers, between which the two sacs are in communication ; whereas the anterior edges are sharp and form a pair of chisels. Not im- probably the infolds bearing these plates may represent the highly modified parapodia of a single segment, opposed to one another by the invagination. No ganglia or considerable nerves are associated with them, at- tributable to this invagination ; and hence its ganglion pair is probably the one which, with its pair of cirri, subsequently advanced to the brain, Perhaps by this time the nerve cords were more ready to free themselves from the hypodermis. Probably the nerve cords were still lateral in position, or at least not closely approximated in the ventral line as now. On our theory the ganglia of the lower jaw segment were the last of the ganglia from the central nerve cords to join the brain, forming mid-brain III, the largest as well as the last of these additions. Like their predecessors, they in their turn addressed them- selves to, and arrived behind, the fore-brain, for they alone of the mid-brain divisions receive the oesophageal connectives; and they, too, pushed their predecessors to the rear. Again, they brought to the fore-brain not only the central nerve cords which in this case persist as the oesophageal connectives with their two roots, but also the ends of a new loop of the visceral nervous system which persist as the pharyngeal nerves, reaching the brain with the ventral roots of the oesophageal connectives. As before, the other end of this loop has been short-circuited and has disappeared. no. 8 THE POLYCHAET—RAW 19 Mid-brain III, the anterior section, does not carry an eye in polychaets. Perhaps it is because by this time the eyes of the head were so much more efficient than the segmental eyes of the trunk that the pair belonging to the lower jaw segment has aborted. On the other hand, it should be noted that the present first body segment is without an eye-spot, so the abortion may have been effected earlier. It will be agreed that the presence of five (the equivalent of six) similar antennae, as is still seen in the Eunicidae and the Amphi- nomidae, is the primitive condition in the polychaet. The development of so great a number by nature on part of a unit brain, such as is conceived either by Racovitza or by Hanstrom, is extremely im- probable, if not indeed unimaginable; whereas, on the theory here advanced, it follows quite naturally. The parallel evolution of the mid-brain, the stomodeum, and the visceral nervous system have now, it is hoped, been read at least in outline. COMPLETION OF THE BRAIN The mid-brain has now been evolved and added to the fore-brain. Arguments have already been advanced for the addition of the hind- brain or nuchal brain last of all. Its incorporation is probably to be attributed to the rearward extension of the brain due to the successive additions of mid-brains I, II and III, which brought mid-brain I to aggregated ganglia of the nuchal organs. This suggests that the incorporation of the hind-brain and of mid-brain III took place at the same time. This completed the basic plan of the polychaet brain. Sodderstrom, however, who first claimed that it was an addition from the body, claimed also that for this reason it could not form part of the prostomium. But had he espoused the theory here advanced, on the same principle the whole of the mid-brain would have to go. Gustafson, though he accepted the first claim of Soderstrom, retained the nuchal brain as part of the brain and presumably of the pros- tomium. On the theory here presented there is still greater reason to accept the hind-brain as an integral part of the brain and of the prostomium. This question is further discussed below (p. 25). SUMMARY OF THE AUTHOR’S THEORY Now that the intimate relationships between these three systems of the brain, the stomodeum, and the visceral nervous system have been deduced and outlined, it is advisable to picture in greater detail the modus operandi of these supposed changes which are illustrated by table 2, stages I to 10. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III As discussed below, we can hypothecate for the primitive annelid the presence of three pairs of longitudinal nerve cords represented now by the nuchal nerves, the podial longitudinal nerves, and the ventral nerve cords. Their positions in the primitive annelid, if symmetrical, would be dorsolateral, lateral, and ventrolateral respec- tively. As the annelid was already segmented, we can call these nerve cords the nuchal, podial, and ventral chains. In the primitive annelid (table 2, stage 1) the primal braim, repre- sented by the present fore-brain, must already have constituted the dominant nerve center of the body; and, judging from the structure of the brain of the polychaet and especially the eunicid, it must have retained its dominance till the brain was completed. The original anterior ends of the ventrolateral nerve cords or “ventral chains” issuing from the primal brain are represented now by the wisceral oesophageal cords, not by the perioesophageal connectives. The in- vaginations may all have been brought about by the animal finding it useful to employ appendages within its alimentary tract and there- fore invaginating them. Further, as its habits changed and the appendages improved, this was effected three times over. This mode of origin seems especially probable in the two pharyngeal invagina- tions where we still have the armature in each case; and if it is accepted for them, it can with great probability be hypothecated also for the oesophageal invagination where all armature is absent. The effects of the stomodeal invaginations upon the primal ventral chains may ultimately be read in detail; but if the podial chains were also involved, we cannot at present cite effects due to them. The oesophageal invagination (table 2, stages 2 to 4). The oesopha- geal nerve loops produced by this (fig. 2) are perhaps represented now only by the oesophageal cords extending from the brain to the supra-oesophageal ganglion (g,) of the visceral nervous system, the upper half of the ganglionlike circumoesophageal ring (co 1) and the continuations rearward in the main lateral pads of the oesophagus (noel, figs. 4 and 5, Heider, 1925, pp. 86-88). Unknown primal “ganglia” along the last are suggested by a few plus signs. The loops were closed by the advance of the foremost pair of uninvaginated ganglia to the brain (fore-brain) to form mid-brain I, bringing with them the ventral sides of the oesophageal loops and the ventral chains (table 2, stage 3). The nerves were brought up to, and appropriated by, the fore-brain, whereas the mid-brain I presided over a pair of 1 But the idea was first suggested to the author by the characters of certain trilobites, where several pairs of appendages seem to be entirely hidden within the alimentary tract. no. 8 THE POLYCHAET—RAW 21 antennae and a pair of eyes. For after the invagination the parapodia of this segment may well have become restricted to a pair of dorsal cirri—‘‘tentacular cirri’—as in the first body segment of Eunice, associated with a pair of eyes, present on all the other body segments ; and both cirri and eyes advanced to the prostomium with their ganglia. The oesophageal loops, though represented in stage 4 as simplified, may perhaps have continued in existence until the simplification of the first pharyngeal, stage 7. The rst pharyngeal invagination (stages 5 to 7). The pair of nerve loops due to this (stage 5) are apparently represented by the lower half of the oesophageal ring and the cords joining gs, g3, and the X-form body, together with part of the pharyngeal cords (n ph of figs. 4 and 5). The unknown primal ganglia along it are again suggested by a few plus signs. The loop was closed (stage 6) by the advance of the foremost uninvaginated ganglia to the brain to form mid-brain II, which also carried with it pharyngeal cords of the visceral nervous system and the ventral chains. But, as after the oesophageal invagination, it was to the fore-brain that they came, entering between this and mid-brain I and pushing the latter to the rear. As yet, there was no fusion between those brain parts. Again the fore-brain appropriated these added nerves, which arrived outside of the previous system. Mid- brain II again, like I, had to preside over the pair of antennae (the previous “‘tentacular cirri’) and the pair of eyes, which had advanced with these ganglia. After this stage came a simplification of this Ist pharyngeal visceral nerve loop, and not improbably of both the oesophageal and Ist pharyngeal loops together (stage 7). If two complete loops still existed, much was redundant, and the dorsal side of the first pharyn- geal seems to have joined the dorsal side of the oesophageal ; whereas other parts aborted. The 2nd pharyngeal invagination. This, the lower jaw invagina- tion (stage 8), may be expected to have produced small nerve loops. No ganglia are associated with the lower jaw, so no ventral ganglion pair was involved. The invagination was probably due to the use by the mouth of the previously foremost pair of extra-stomodeal appendages, the dorsal cirri of which at the same time functioned as tentacular cirri! These appendages are represented by the lower jaw plates and pads. The new visceral nerve loop was closed by the advance of the fore- most pair of ventral-chain ganglia to the brain, in the same way as before, to form mid-brain III (stage 9). With them to the prosto- 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL irr mium traveled the tentacular cirri of the time to form the antero- lateral tentacles. No eyes accompanied this third pair of cirri; and it has been mentioned that eye-spots have been lost from the first body segment (that with tentacular cirri) in Ewmce punctata, The added ganglia, following the established rule, addressed them- selves to the fore-brain, as had their predecessors, whom they in turn forced to the rear. They trailed again the ventral cords and by applying themselves to the fore-brain caused the double roots of the oesophageal connectives. They brought also with them new visceral nerves—the present fore-brain roots of the pharyngeal visceral cords. It is significant that these arise from the ventral roots of the oesopha- geal connectives—both were brought together, and at the time of the addition of mid-brain III. How exactly the simplification of this visceral system was effected may not be clear; but the three courses of stage g imply redundancy, and one course alone seems to have survived, namely, the most ventral, because this alone would be associated in origin with the new oesophageal connectives. One other change and the basic plan of the brain was completed: mid-brain I had now been pushed so far to the rear that it has approached the nuchal ganglia and these also have added themselves in stage 10 to constitute the hind-brain. With this the brain of the ancestral polychaet has been evolved. The stomodeum is completed, and this has involved the completion of the brain. | LIMITS AND SIGNIFICANCE OF THE PROSTOMIUM OF THE POLYCHAET Regarding the extent of the prostomium, there has been much difference of opinion: it has indeed been a subject of keen contro- versy. But there is perhaps no need to recall here the numerous divergent views. Considering the great antiquity of the polychaets, it is not surprising that the head should present great variety, and suggest different opinions. The ideal prostomium, the prototype, is the region in front of the mouth in the common ancestor of the group; and this may be expected to be indicated by what the most primitive polychaets have in common there. By previous writers this seems generally to have been equated with the prostomium of the hypo- thetical primitive annelid; but this is opposed by the claims advanced in this paper. As to its nature and origin, two markedly contrasted conceptions have been held: (1) that in its origin it was segmental in some way or other; (2) that it was a unit, and unsegmented (see table 1, No. 8 THE POLYCHAET—RAW 23 p. 4). Soderstrom, who held the latter view, sought in effect to change its definition into the unsegmented anterior region of the body; and he cast scorn on all who had not the same “prostomium idea.” Such a definition is, however, quite unjustifiable. The conclusion of Binard and Jeener, supported also by Gustafson and accepted by the present writer, is that the primitive appendages of the head comprise an anterior pair of antennae, modified in some forms into palps; a second pair, never so modified; and a median antenna. These are innervated by successive parts of the mid-brain. The morphological order of these appendages was naturally of less importance to those who regarded the prostomium as a unit, than to those who thought of it as segmented. The earlier writers, previous to Lameere, all placed them in their order of position—the anterior pair first, the median antenna last. Lameere reversed the order, and was supported by Binard and Jeener. Hanstr6m seems to have accepted the reversal, but did not accept Lameere’s theory. Gustafson was apparently in doubt, and perhaps for that reason numbered them chaotically. The present theory erected in entire ignorance of that of Lameere, also reverses the order, but for an entirely different reason; moreover, whereas according to Lameere the sequence of the three “pairs” was already developed in the coelenterate ancestor, on the present theory the three pairs of ganglia that innervated them (as well as the antennae themselves) were picked at random, so to speak, by nature out of a long sequence of postcephalic segments during the evolution of the polychaet, and carried forward to the head. Besides five antennae, the prostomium bears also two pairs of eyes. Even those who have recognized segmentation in the prosto- mium have not regarded the eyes as segmental ; but the author claims that they belong with the intermediate antennae and the median antenna, and are thus segmental in origin (pp. 16, 18, 21, and 22). Previous to the researches of Soderstrom on the Spionidae the prostomium was always considered to include also the whole of the compact mass of nervous matter constituting the brain. He, how- ever, homologizing the nuchal organs with the dorsal, segmental, chemical sense organs of the Spionidae, which extend through the whole body, claimed that the hind-brain and nuchal organs must therefore be excluded from the prostomium. On the other hand, all the most primitive polychaets have a hind-brain closely associated with the remainder, and possess also these nuchal organs; hence there is every reason to credit the ancestral polychaet with a hind-brain and nuchal organs, and therefore to include these parts in the pro- stomium. And just as the nuchal organs have their homologues on the 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III postcephalic segments of the Spionidae, so the antennae have theirs in the dorsal cirri of the body segments, and the eyes have theirs in the postcephalic eye-spots. The principle on which Soderstrom would exclude the hind-brain is invalid, and is reducible to an absurdity. On the theory here advanced it would also exclude the whole of the mid-brain, the antennae, and the eyes. Furthermore, reason has above been given for the incorporation of the hind-brain at the same time as mid-brain III. Again, on the theory advanced by Hanstrom it would exclude the fore-brain and the mouth-lips! The principle is wrong: Séderstrém did not allow for any evolution within the annelid. We must get back to the primitive idea, and define the prostomium as all that region in front of the mouth inherited in that position from the primitive polychaet. On the theory here advanced this includes: (1) the fore-brain and the representative of the “mouth- lips,” inherited from the head of the more primitive annelid, to- gether with the following parts acquired in order by the prostomium during the evolution of the head of the polychaet: (2) the hind sec- tion of the mid-brain (mid-brain I) with the median antenna (due to coalescence of a pair), and a pair of eyes (the posterior pair) ; and long afterward (3) the middle section of the mid-brain (mid- brain II) with the second pair of antennae, and the anterior pair of eyes; and again long afterward (4) the anterior section of the mid- brain (mid-brain III) with another pair, the anterior, of antennae; and also (5) the hind-brain with the nuchal organs; these two brain parts (mid-brain III and the hind-brain) being probably acquired nearer the same time. Gustafson, who is doubtful whether to accept the theory of Hanstrom that the stomatogastric lobes (or fore-brain) have been added from the ventral chain, is doubtful therefore whether to regard the mouth-lips, which they innervate, as part of the prostomium. Nothing better illustrates the enormous contrast between the present theory and that. What Hanstrom regards as the last addition is here claimed as the original nucleus. Many may be surprised at the inclusion here of the hind-brain after Soderstrom’s important and significant work; but it will be clear that the same principle, which would exclude it, would exclude also the whole of the mid-brain and all the cephalic tentacles and eyes. It will thus be seen that the theory here advanced presents a new conception of the prostomium. It is not the representative in modern forms of the head of the primitive annelid, but the result of the long evolution from this of the head of the polychaet. It is not a unit as so many have supposed, but an aggregate, acquired only in the No. 8 THE POLYCHAET—RAW 25 course of unimaginable time. It is not due merely to the cephaliza- tion of additional anterior segments of the body as many have sup- posed, but is largely due to the advance to the brain of three pairs of segmental body ganglia at three far separated times, and, at the same times also, advance to the head of three segmental pairs of cirri and two segmental pairs of eyes, to become the cephalic sense organs. Incidentally this aggregation caused the brain to incorporate also the hind-brain innervating the nuchal organs, which are therefore just as much parts of the prostomium as are the other cephalic sense organs and the mid-brain. Further, this great aggregation of the brain and the cephalic sense organs, which built up the polychaet prostomium, is claimed here as incidental to the evolution of the stomodeum, which again is referred to changes in the mode of feeding. MORPHOLOGICAL SIGNIFICANCE OF THE HIND-BRAIN The nuchal organs are quite dorsal in position. This is seen very clearly in the Spionidae which in some forms retain them completely ; and it is indicated also by their position in the errant polychaets. But in these it is only in the tetraneural Amphinomidae and Euphro- synidae that extensive development of the nuchal organs occurs. Here development of the caruncle is accompanied both by great enlargement of the hind-brain (presumably by the addition of their ganglia), and by tetraneury; and again the nuchal organs are quite dorsad of these nerves. These relationships suggest that the ancestor of the polychaet, before the evolution of its brain, stomodeum, and visceral nervous system was characterized by sexneury having: (1) a pair of ventral nerve cords (still persisting as the ventral nerve cords of the central nervous system), (2) a pair of lateral or podial nerve cords (still retained only in the tetraneural families above men- tioned), and (3) a pair of dorsal or nuchal nerve cords (still retained by the Spionidae, greatly developed and cephalized by the tetraneural families, and greatly reduced subsequently in other families). The mode of evolution of the polychaet detailed above—by three ventral invaginations (two great and one small)—would involve, also, as these took place, the contraction and packing together of the dorsal parts of the ventrally invaginated segments and, therefore, of the nuchal ganglia. The tetraneural families, as they retained their podial longitudinal nerves, took advantage also of this, and have incorporated the nuchal organs into the caruncle, and the nuchal ganglia into the brain. 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS, Tit MORPHOLOGICAL SIGNIFICANCE AND RELATIONSHIPS OF THE PERISTOME On our theory the peristome is not a somite but marks the site of a succession of breaks, causing a gap in an originally continuous sequence of ganglia, the remainder of which form the ventral nerve cord. Further, in the course of evolution the peristome has been extensively exchanged three times over. The missing members are believed to be partly distributed along the visceral nervous system, and partly coalesced with the brain. The magnitude of this “gap” furnishes a complete explanation of the hitherto mysterious fact— that the nervous system arises from two distinct rudiments, one giving rise to the brain, the other to the ventral chain—a fact which has been noted by numerous workers from Salensky (1882) and Goette (1882) onward. The visceral ganglia are thus brought into the succession of those of the central nervous system, in opposition to the generally accepted view of their separate origin. With more detailed knowledge of the visceral nervous system it may be possible to determine in large degree the succession of the various ganglia which can be recognized. In the brain, besides the fore-brain and hind-brain we have an association of three pairs of comparable ganglia, sending out com- parable pairs of nerves, and innervating comparable organs; and according to this theory they were added to the brain at three different times. Further, they were selected by nature in the course of evolu- tion from a considerable length of the primitive ventrolateral nerve cords, including probably well over a dozen pairs of ganglia. The order, too, of the upbuild of the brain on this theory is quite different from that on any other. The contrast between this plan and the others that have been advanced could hardly be greater, whether we com- pare it with that of Pruvot, or Racovitza, Nilsson, Lameere, or Hanstrom. The reason for this is that it connects the three systems— the brain, the visceral nervous system, and the stomodeum. It is curious that the order of the elements of the mid-brain happens to be the same as on Lameere’s theory; but the significance is entirely different. It may be remarked that there is much to be said in favor of Lameere’s theory as a mode of origin of the annelid; not, however, of the polychaet but of its distant ancestor. Should our theory prove correct, Racovitza will have builded wiser than he knew, though quite other than what he meant ; for the primal brain was his fore-brain, to this was added his mid-brain, and to this the hind-brain. But the evolution of the head and brain envisaged here is entirely no. 8 THE POLYCHAET—RAW 27 different from that of Racovitza. In his view the peristome did not differ fundamentally from a normal segment (1896, p. 154); the head of the ancestor became the head of the polychaet; the parts were already there, including the primal elements of the cerebral nervous system—the “‘aire palpaire,” “aire sincipitale,’ and “‘aire nucale.” Only further development of these was needed: that of the sense organs on them into palps, antennae, eyes, and nuchal organs; that of the areas themselves into the fore-, mid-, and hind- brains. Before this evolution in the polychaet there were no com- parable sensory organs (1896, p. 161). On the contrary, according to the theory here presented, the ancestor was already furnished with eyes and parapodia throughout the body, and possibly also some form of appendages and eyes on the head, which already possessed a brain and bore also an early stage of the mouth-lips. The 4 (or more) pairs of ganglia, too, which initiated the mid- and hind-brains were already functioning in their own postcephalic segments. But whatever sense organs that primitive annelid had on its head, except for the mouth- lips these organs were superseded in the evolution of the polychaet by previously postcephalic eyes and antennae; while its brain was to be extensively supplemented and partially superseded. A fundamental difference between the two views is that Racovitza in evolving the polychaet brain thought he was evolving that of the primitive annelid, whereas on the here offered theory no primitive annelids survive. MPPLICATION OF THE AUTHOR'S THEORY TO POLYCHAETS IN GENERAL Very close comparisons are now possible between the different families of errant polychaets in regard to the prostomium and its appendages and eyes, the brain, the stomodeum, and the central and visceral nervous system; and, to judge from the results of past re- search, the suggestions of close affinity are not likely to be diminished in the future. In the accompanying table typical representative sub- orders and families are tabulated in regard to their brain, sense organs, stomodeum, and buccal segments; in it, too, the homologies are indicated, and characterized as succinctly as possible. The agree- ment in brain and sense organs is extremely close; and even in the stomodeum, though the characteristics vary so much, a series of stages can be seen between the Eunicimorph through the Amphino- morph to even the Glycerimorph. In table 3 the order of development of the different sections of the stomodeum is to some extent indicated by the letters a, b, c; and the order of incorporation of the divisions of the brain by A; BI, Bil, BIE; C. 28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III So close are these comparisons that one can have little doubt but that all derive from the same ancestral stock; and if the fundamental plan of the brain above detailed be accepted, it will be agreed that the closest approach to this ancestral stock is to be found in the Eunicimorpha. The evolution which has been traced, if it be sub- stantially true, is therefore that of the ancestral polychaet. It can all be assigned to very far back in pre-Cambrian time, effected say before 1,000 million years ago. Of this the writer feels confident ; because, as he hopes to show, the arthropods, some of the classes of which were evolved long before Cambrian time, were derived from a particular family of polychaet still extant! If the above given conclusions as to the character and origin of the original polychaet brain be accepted, its further evolution can with some confidence be pictured. The separate ganglia of the two sides of the brain united in the midline, consolidation was effected, and enormous development has ensued—development which can be read by detailed comparisons of the brains of the various polychaets. The fore-brain, hitherto dominant, has lost its old eminence even in the Eunicidae; and in other families is often greatly reduced. The mid-brain has undergone great development, and perhaps in all forms has become the chief nervous center. And whereas, as is here assumed, the palpal function in the primitive polychaet was seated in the mouth-lips, and served by the fore-brain, now only in the Eunicidae and Amphinomidae is this the case. In some others it is apparently seated in the proboscis and is still served by the fore-brain: in yet others (viz, the Aphroditidae, Chrysopetalidae, Hesinoidae, Syllidae, and Nereidae) the function has been transferred to the anterior pair of antennae, transformed into palps, and is served mainly by the mid-brain. This is indeed the case even in a family of the Eunicimorpha—the Stauronereidae (Staurocephalidae). The degrees of development of the brain are extremely varied; and the new structures include neurone courses, nerves, commissures (trans- verse in considerable number, and also longitudinal), the true brain ganglia, and in the highest forms the corpora pedunculata. In the Amphinomorpha extra pairs of nuchal ganglia have joined the hind- brain and this has here become much the most voluminous division of the brain. With different modes of life very different lines of evolution were followed. All other suborders seem to have changed more than did the eunicid. Here the brain is still relatively primitive, exhibiting its primitive subdivisions still extended in a plane. In contrast, in the aphroditid it is compact, folded over between front and rear and very highly developed. On the other hand, in some no. 8 THE POLYCHAET—RAW 29 errant families and perhaps in all the sedentaries it has degenerated. Regarding the stomodeum, if, as is above suggested, the much- folded form exhibited by the Eunicimorph is the most primitive type, the different errant polychaets present us with suggestions of the stages of evolution between this relatively nonprotrusible folded type and the extremely protrusible cylindrical type of say the Glyceri- morpha. Such an evolution one might expect to be accompanied by great changes in the visceral nervous system, and probably also in the brain. To trace the evolution of any form and the deployment within the class, it will be of great advantage to have a starting point, a basic plan, which is all that the present paper can claim to attempt. Were zoologists to attempt to work out by ontogeny and by comparative anatomy the many courses of evolution within the class, the common starting point would soon be apparent. uoisn1ijo1d SuIMO][e ‘pazeo “yd A[3u01}s *Q[911JUIA ito snseydoss0 ‘uasyoiq Ajjuated||ArtjaururAs [e1aje[Ig| Jejnosnur “ve } -de Suleq ysiy 94} ‘Wt puryeq 4ys1y}| seyjided yyM yng) Je[Npuels -q 94} SI VINSSIWIWIOD [L1ZU2A PUOdDS 91{}||*Y}22} 10 smef ON xuAieyd (q) x Ajjensn pue fuoijeisiu paeMivei pure edi 0} Ie} } 9ZIS JO oseaIONUI }eoI1s patoayns sey SuIyoest pri} sulusdo yjnouw 9y4} Awe} oY} UIyWM -udsA yng ‘xu wg <“yH puryeq saANsstuiuI0d 4194} YUM PIjsues [e1}USA }SIY ay} Sey pue auloysiied 94} Aq papunosins st yjnow oY} ‘amourygupsvg Ul se ‘A[VATPIWIAgG ‘eipodeied sur1esq euou IO 11119 1e[n -9}00} jo 11ed sug quoulges 1 sosepuodde on quow3eas 1 pue ‘paz}ejusi1syip dUON 913311 aUl0}ST19 -Ieyd 94} 4}IM pesiop snon -ulqU0o9 ‘Y}eaYs a[qistaaa (9) sosepuadde on asie] AZ A sosepuodde ony Bie] 194}ey uoIs -najoid 9}e1Npoul yy s1osoqoig yso] AIJOUIWIAS datliivs ‘suey sojejd Mes z sn3 soyeid qni z -eydos30 suo] (e) apis 7y4b14 Jow.1e Jatiies ‘suey [eorsjeuuAs sajeid mes £ ‘uns A[ [8.1932 sajejd qni z “Iq YUM ‘yORs apis gfay (q)|Mel toeddn (q) soyejd jo sared Axjou -wAS [e19}e]IG pesied |e ‘opis yor uo 4399} «= snodour “nu AI9A JO Sart -9S d10UI 10 Zz espa 2a1} 72 poyy00} ‘sazejd jo sed (9) (q) (2) )yoers Mel amo] (9) uoisniy4 -o1d =poiumiy, A19/\ ureiq pury 94} fo eljsues [e4eA0s Aq podAtos ‘s}UsWI39S “ue J9AO0 SUIpUus}xa apounies OUT JUS -dojaaeap eweidnsg aulojsiiad jo JuoIT Jepun sued 909 sn3 -eydos9o0 Suo[, (®) [Bors -youumAS = AT[B -19je]Iq ‘*yoes Mel aaddn (q) yores Mef 1aMo] (9) uoIsn.t4 -O1d poy s19A syd z our paprarp -qns yore ‘SMO1Ing) jeyonu jo ted sue xudivyd jo 10w1y sjuowm3gos ]eoong Arossa00y PWO STII 119308 -1849 PUP SUOISIAI(T umnapoui03ig supv640 JDYINU 3Yy} BuI}eA -19uUL UleIg-pur}y 0) Avot : UINIUIO}sSOIg (ques IOII9}|-Ge2 10) sake [[euiG]10) seka as1eT 9[0RJU9} URIPIPW diel SIAT a[9e}U9} URIPIT quas “qe 10 safkd [[eWsS juesqe 10 ([[I119A sdoupjau *S ‘aqniry DAQNA “S) [BISIISIA 9[9e}U9} uPIpPIy saf2 jo ied & pue (ated paosa]e20o eepimouryduy VHdYOWONIHANY (2eprorpoaT) eepioiuny (sny -eqdao0ineys) SI91ZUOINeS (2epr “o[FA10q = aept (Juasqe (eZ}IAOIeY JO] (ez}WAOCDeY jo sdjed jo" oapo||1oydodjed) sdiy $9[9B}U9}|-4}4S) safoejus}|-yjynou 10119} e}eIpeuitajuy| [e19}e[01}U9 A |-ue puke [e19}e'7 soAq (uoszejsn5|,.Sdjed,, peje sojoeque}|fo sdjed) sajoez|-OS *sdij-yynour d}eIpauiiszuy|-us} = BpsjnG|AOF4te}Ue Fo Areg sofa as1e7] suesi0 ,odjed,,se pedo|pauyap ou 4M sopeuoy|-[aaesp Ajjeain|sdiy Jody saAo jo ared e& pue $319D4JUIq S3jID4Udy SGtf 401434 Be) 2j9Djua, UDIpam|yjo ied e Bur\jo aed eB Bul|-wp 94} Ali1eur 243 SUIJLAIOUUT|-JLAIQUUL UlTeIg)-}eAIoUUT UleIq)-l1d suljeAroU UIVIG-PIUI IOII9}SOg/-pIul UPIPey|-plut AOMo}UY|-Ul Uleiq-310,j Id Ila Ilia \"f umsiop ‘um s anne Pp :uintu0}so1g pue ju0sy : WINTWIO SOIT -[eydao0ine3S) PEpla1suOINe}S VHduONIDINA Soy Ey poe siapioqns LAVHIATIOG SUOISIAID ureiq Surpuodsaii07) umorge1 Peay JO SUOISIAIGD (payyimo saymuvf s6njag¢) sjavyakjog yuvasa yooigay fo uorhas poay ayy fo sarrgzua porbozoyguow ayy fo sanbojouoy pasoddns fo uostapduoj—€ atav ], (1£) me Iida Ieynoejue} jo sired Zz YM yqoea ‘gui0jstiad 243 YUM poaos2] “P09 10 ‘ajer1edas oq Aevul oejas jnoO “WIM sjusuIsas £-0 yore jo aseq jv sayy “Sliq jo o2[pung Be YUM ‘111d Ie] -norjyua} jo sired Z YUM IUIO,SIIIgG 2e}9S YUM eipodeied 1edq AvW ADYL “i419 Jejnoejus} Jo (je1u9A pure Jesiop) ssied < Buliveq yord ‘juau “BS 4X9U YIM pa0sa[BOd suUIOSIIOg Aljow [eoripurpséo [e193] 1g smef Auiog yoy} Jo YyM pue ‘oeqided yum (q) -wAS ysoulle AljawUIAS [e19}LIG puryaq saspit ayposnur pue ‘quo1y ur smef ASIBASUBI} (syjeuseired) yq99} YUM pue aeyided = ym (9) (q) LLIID ON qustses }yxoU YIM pasny dsul0}sTI97 11119 1ep]ndejU93 A[UO ivoq Aew sjusuZes 419430 OM} IO 3UQ §}usuI 39s yeoonq Arossa.0y 11119 1epNdvyUs} jo sired OM} 10 au0 YIM dW0}SII9q dUIOIST19g AijourwAs [e19} e/g auou 10 sMef Ausoy ys0ys z (q) xudieyd sututof -pe pue ‘soli -9s = [eUIpNys -uo] ul aeyyided (9) Aajowmusés [eoripurjAd = ysowyy syjeuseied sowmijeuos pue eeyjided = yyM (9) nowy eB09ed jo ed & YUM ‘apolizueA (e) xuAieyd dejnosnu (q) yyeays a[qIstaaad (9) stosoqoid a[qisnajoid aiqis -U9}X9 AIBA ‘BU0ILS BO9ed Z WM = 9]9T1}U9A to snseydoss0 (v) sa]o -SnUI 9AISSEUI qyM xucreyd (q) yyeoys avjno -SnUL 9I[qISI9Aa (9) srosoqoid aqisnajo1d §=38u0135 apoIIJUGA (ev) sasej4s oM}ulxuAIeyd (q) yyeoys a[qIstaAa (9) stosoqo.id aiqisnajoid 4104S BI9BD Z YIM Ud}JO B[OIIJUIA (v) suorjoas z ul xudieyd aqqisnajzoid (q) yyeays aqisn1} -o1d = ajqisiaaa (9) sto -soqoid 93] qIsn1j01g UWINIpouloj}y suesi0 yeyonu say g sues.10 [eyonN (Sggr ‘j0A “Nig 0} surIp10s90v) quesslg suesi0 Jeqgonu dsATWUIIIgG soAq (quasqe 10) 2[9eJUI} ULIPS|W So49 IO[TBWS quosqy yuesqy quasqy (juesqe 10) soxy (quasqe 10) 9[9e}U9} UPIPIT uresq-puly] Ue CSE IN paenuyuoj—t aIav sox (auotsa yy] (juasqe 10) |\ur A[UO JUasqe) sus sopoejuay|pezyuiol-z sdjeg}|-10 pauyesp oN soko o51e'T S}uI0f,, ——$—— & WM yors sues sajoejua y |sdjed asiey]|-10 pauyop ON sof (sot119}s0d) (401194 sues sopoejuay|-ue) sajoejuay|-10 pauyop ON SoA sues sojoejua s9[oejuay|-10 pouyep ON IT Uresq-pHW | TIT Bresq-pHy | wresq-240,7 oepruoiseH aeplsieny aepipAyiyden aeproopol| Aud VHdutOW ada NT Sol[Imey pue siop -1loqns jJaeqokjog (z£) ce RRR“ sosepuadde DUON|OU YIM sUIO}SIIIg aseq aq} Buoj|ye 9eyIS YIM T1119 AIA JUSUIZES }xX9U|1e[NIeWUI} Jo siied JO SMI [PIJUZA|Z YUM UWO}SIIOgG avyas ou ynq lito) ievjnoe} -U3} JO Sited z 10 SUON|I YIM sWIOISIIIg sjusuges PUIO}SIIOg JeBoong A1osse.0y AayowMWwAS [BolIpuljAd = jsowlyy syj}eu a[olJUaA (ke) -Seied snoi9w xuAieyd -nu pue sme z aiqisn14yz so smef + yyIM (q)|-o1d se[Mosnu (q) aejided qyeoys peseaod (2) a[qisteaad (9) srosoqgoid 9]qIs -najoid-1e} 9}eAvIO 4O [eolipul[Ad 2uo0y yaa AIJIWIUWAS [BIDE SME ][V1ZUIA FO ared & pue [es -Iop jo tied & V[oIsjUaA (Be) pue sejjided jo xudieyd ear eB yum (q)| —a]qysn4jzo1d (q) yjeeys a[qistaaa (9) stosoqgoid 92]qIs -nijoid [eo1ipurjAd Le} 8) jo aed & yno UELEES ef) UMA a]oraqueAjsod (Te) AI}OUL a[sriyueA (Fe) -wiAs Jeoripurféo} = ayorayuaaoad ysoulye ‘je19}e]1gq| W4oj-jat1eq (fe) ovyided Aq pemol YUM pue ‘]e1o -]O} xudieyd -AIS IO Y}00} TeorapurpAo jesiop 1 yyWM (q) aqisnajoid (q) yyeeys 2IqIsiagAa (9) srosoqoid e]qisnajo1d suoy inow1y WINnepowi0}S uesio [eyonN uesio [ByOnN Iggi ‘Jasueds 0} Surpioos.e jue -said uesio [eyonyy uyesq-pulyy (4z61 ‘wots -Uv}{) “WN1UI0}s yuesqe Ajjensn -oid poye[nuue yussqy|1o seta ON Ajliepuooas [eo -1U09 94} JO 194 UWINIUI0}s|UNTWIO}sOId Jo|-uaA ay} Aq po -oid jo puod }e/pua ye sajoe}|/-juesoida1 ‘poy quasq VY | sajorj}ue} todd ¢) |-ua4 J9MO'T|-poul A[auI9 14 XG SOAY SsoAq sues sdjed suoy|-10 pouyop ON (Juasqe 10) $3]0e}U9 T, (juasqe 10) 9[9ejU9} URIPIT (121 Wor} uMBIpP jJuesaiId aq Aeut sired vijxo z) soAq sefq| poajyurof yonur ‘sdjed padojea sues saoejUa Ty |-2p Aj}eai5H|-10 peuyosp ON II Uresq-PUN | IIT Urezq-pH | Uresq-210, 9[9jU9} ULIPIPL J wreszq-PIW panuyuoj—t adv J, SEPIIad4[5 VHdaYOW THAD ATS) evuloudjog avdILiaowHay SePHIAS sorplmey pue slop -1oqns jeeyoIAjOog no. 8 THE POLYCHAET—RAW 33 CONCLUSION Both in their complexly folded stomodeum and in their brain the Eunicimorpha are claimed as the most primitive of polychaets. Their study suggests the characters of the ancestral annelid and the subse- quent evolution of the complexities of the stomodeum, the brain, and the visceral nervous system of the polychaet. In brief, on this view, the ancestors of the polychaet, in connec- tion with their successive modes of feeding, transferred not only the external skin, but also the nerves and ganglia of the central nervous system, which at that time were lateral in position, and also the appendages associated with those ganglia, up the alimentary tract in the formation of the stomodeum. As a primary result they have originated (or greatly added to) their visceral nervous system; and as a secondary result have complicated their brain. The dominating factors were the invaginations ; and the structure of the eunicid shows that these took place three times over. (An analogous process took place yet again in the evolution of arthropods, as the writer hopes to show.) Most of the fundamental ideas expressed in this paper are already summarized on pp. 19 to 27. Though the evolution here claimed is so great and so widespread, it is not believed to be beyond the capacities of variation and natural selection to effect. As a last word the author desires to thank Dr. E. G. Butler, Dr. J. Percy Moore, and Dr. Ernst Mayr for their encouragement. REFERENCES Batrour, F. M. 1881. Comparative embryology. BERNERT, J. 1926. Untersuchungen tiber das Zentral-nervensystem der Hermione hys- trix. Zeitschr. Morph. Okol. Tiere, vol. 6. Berlin. Binarp, A., and JEENER, R. 1928. Morphologie du lobe préoral des Polychétes. Recueil Inst. Zool. Torley-Rousseau XXX, vol. 11, fasc. 1. Bruxelles. Etter, E. R. 1945. Scolecodonts from the Trenton series (Ordovician) of Ontario, Quebec, and New York. Ann. Carnegie Mus., vol. 30, pp. 119-212. FAvuvEL, P. 1923. Polychétes errantes. Faune de France, vol. 5. Paris. GuSTAFSON, G. 1930. Anatomische Studien itber den Polychaten Familien Amphinomidae und Euphrosynidae. Zool. Bidrag, Uppsala, vol. 12. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, III HAMAKER, J. 1808. The nervous system of Nereis virens. Bull. Mus. Harvard Coll., vol. 32. Cambridge, U.S.A. Hanstrom, B. 1927. Das zentrale und periphere Nerven-system des Kopflappens emiger Polychaten. Zeitschr. Morph. Okol. Tiere, vol. 7. Berlin. 1928. Vergleichende Anatomie des Nervensystems der wirbellosen Tiere, etc. Berlin. HatTscHEK, B. 1891. Lehrbuch der Zoologie. Jena. Herper, K. 1925. Uber Eunice. Systematisches, Kiefersack, Nervensystem. Zeitschr. Wiss. Zool., vol. 125. Leipzig. HEMPLEMANN, F. 1911. Zur Naturgeschichte von Nereis dumerilii. Zoologica, vol. 25, No. 62. Stuttgart. 1934. Polycheta, in Kitkenthal-Krumbach, Handb. Zool., vol. 2, 2d half. Henry, Laura M. 1947. The nervous system and the segmentation of the head in the Annulata. Microentomology. Nat. Hist. Mus. Stanford Univ., Calif., vol. 12, pt. 3, pp. 65-82; pt. 4, pp. 83-110. Hormcren, N. 1916. Zur vergleichenden Anatomie des Gehirns von Polychaten, Onycho- phoren, etc. Kungl. Svenska Vetenskaps Ak. Handl., vol. 56. Stockholm. KLEINENBERG, N. 1878. The development of the earthworm Lumbricus trapezoides. Quart. Journ. Micr. Sci., vol. 19, 1879. (Sullo sviluppo del Lumbricus trapesoides, Napoli, 1878.) 1886. Die Entstehung des Annelids aus der Larve von Lopardorhynchus. Zietschr. Wiss. Zool., vol. 44. Leipzig. LAMEERE, A. 1925. Les organes segmentaires des Polychétes. Ann. Soc. Zool. Belgique, Lie BAVA TS LAncE, F. W. 1947. Annelideos poliquetas dos Folhelos Devonianos do Parana. Arq. Mus. Paranaense, vol. 6, art. 5, pp. 161-230. Ma taguin, A. 1893. Recherches sur les Syllidiens. Lille. Meyer, E. 1901. Studien tiber den Képerbau der Anneliden. Mitt. Staz. Zool. Neapel, vol. 14. Berlin. Nitsson, D. 1912. Beitrage zur Kenntnis des Nervensystems der Polychaten. Zool. Bidrag, Uppsala, vol. 1. Pruvot, G. 1885. Recherches anatomique et morphologique sur la ‘systéme nerveaux des Annélides Polychétes. Arch. Zool. Expér. et Gén., ser. 2, vol. 3. Paris. no. 8 THE POLYCHAET—RAW 35 Pruvot, G., and Racovitza, E. G. 1895. Matériaux pour la Faune des Annélides de Banyul. Arch. Zool. Expér. et Gén., ser. 3, vol. 4. Paris. Racovitza, E. G. 1896. Le lobe céphalique et l’encéphale des Annélides Polychétes. Arch. Zool. Expér. et Gén., ser. 3, vol. 4. Paris. RopHE, F. 1887. Histologische Untersuchungen tiber das Nervensystem der Poly- chaten. Zool. Beitr. (v. A. Schneider), vol. 2. Breslau. SEDGWICK, A. 1884. On the origin of metameric segmentation. Quart. Journ. Micr. Sci., n.s., vol. 24. 1910. Articles “Peripatus” and “Embryology” in Encycl. Brit., 11th ed. SOpERsTROM, A. 1920. Studien iiber die Polychatenfamilie Spionidae. Inaugural disserta- tion. Uppsala. 1923. Uber die Zunahme der dorsalen Sinnesorgane bei Nerine fuliginosa, etc. Zool. Bidrag, Uppsala, vol. 4. 1927. Uber segmental wiederholte “Nuchal-organe” bei Polychaten. Zool. Bidrag, Uppsala, vol. 12. SPENGEL, J. W. 1881. Oligognathus bonelliae, eine schmarotzende Eunice. Mitt. Staz. Zool. Neapel, vol. 3. STANNIus, H. 1831. Uber den inneren Bau der Amphinome rostrata. Isis. StorcH, O. ‘ 1913. Vergleichend-anatomische Polychiaten-studien. Sitz. Ber. Math. Na- turwiss. K1., Ak. Wiss., vol. 122, pt. 1. Wien. i m MI nyt Oh oe cgay , gl wit, ul: ar , avn bu eA wih s Ay Oth \ ‘ bi 4 1 A u £ ? mt i i wan a ie tae ee yu ‘ ea ; mes fe Ales i ASE 7 bb iy a Re 1 : a a) ve 4 t M ie (ty are if { 1 ) i ee wy Mi Ue 4 / } ary z; ah 1 “1 cal n i . i i J COW ha ; par Lge ne ° (0 Rae 1 eN \ iy ~ . wey “ey ies <\ ot a Ohl joey Lavi « i 4 y ae . ww! iy Te ES “ws as an “ iy 4 : " wee i om bt 1 j i D Y ) 5 } i) : ae | : 1 q 4 1 i Wy ui tf " ’ ; ] ‘ j iii ; i ' at t) 1 i i wt I i ey 1 ! ht i i i . 1 r i wy i . y ) i : , : i 1 : : i t i : : f ; } \. 1 j - { f r it ; Ps ler . i ih : ee 7 ay é | . —, un m ‘3 - a - =a ’ z - : =i 1 = : _ \ 1 B i - : f : hi ; : . Ny . } | f le i # . z. i, 1 ’ A i i \ 1 ' ‘ i 4 Ld } a it a | ‘ ; ‘ ' p| 1 i 1 ‘ » ve - at a. ve vs } + ‘ : AN a u ; VY an : d tp t ry ‘ oo : { oy : . : , ra ; hed i ; a a ae i ; i. ’ 4 uh i Opie. ; Le he 5 i* 7 iv i 7 t I vl fad ny } F ' ; 5 a i _ : hey }) oan t ne ; ) i } ul ¥ a\ 7 a) i i : i ii 7 : a ‘ le | ii ' it a 1 = i i Ne i € [eye i ¥ sth Nees whl vik elf ie | ? a ape SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 9 A NEW HERON AND A NEW OWL FROM VENEZUELA BY HERBERT FRIEDMANN Curator, Division of Birds U. S. National Museum (PusiicaTion 3985) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION JULY 21, 1949 The Lord Baltimore Press BALTIMORE, MD., U. 8 A. A NEW HERON AND A NEW OWL FROM VENEZUELA By HERBERT FRIEDMANN Curator, Division of Birds, U. S. National Musewm Among a collection of birds from northeastern Venezuela is a specimen of an adult male Syrigma sibilatrix, the whistling heron. At first glance this bird was seen to be strikingly different from a good series from southern Brazil, Paraguay, and Argentina, especially in the coloration of the upper wing coverts, but also of the crown, neck, breast, and, less obviously, in size. As no name appears to be available for this bird, I take pleasure in naming it for the collector. SYRIGMA SIBILATRIX FOSTERSMITHI, new subspecies Type.—U.S.N.M. 406385, ad. ¢, Caicara, Monagas, Venezuela, July 24, 1948, collected by Foster D. Smith, Jr. (orig. No. 143). Subspecific characters Similar to the nominate race but differs in having the upper wing coverts much more yellowish and more nar- rowly striped with black—between chamois and honey yellow, not dull pinkish cinnamon to light ochraceous salmon as in the southern birds ; the neck and breast more yellow, less olivaceous—light honey yellow, not light buffy olive as in typical sibilatrix ; crown somewhat paler—slate color (blackish slate in the nominate race) and with the bill longer, culmen 74.1 mm. as against 66-71 mm. in southern birds. Measurements of type——Wing 280, tail 105, culmen from base 74.1, tarsus 88.2 mm. Range.—I have seen only the type, from Caicara, Monagas, but Hellmayr and Conover (Cat. Birds Amer., pt. 1, No. 2, p. 207, 1948), state that the species (undoubtedly this race) occurs in the valleys of the Orinoco and Apure Rivers in Venezuela. In a footnote they write that it “. .. remains to be ascertained by the study of an adequate series whether Venezuelan birds are really quite the same [as birds from southern Brazil south to Argentina]. A single adult from the Rio Apure is slightly larger with longer tarsus and bill, while the upper wing coverts are more predominately yellowish and more narrowly streaked with blackish. The divergencies are, however, insignificant.” How they could call a striking color difference insignifi- SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 9 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III cant is beside the point; their Rio Apure bird is obviously of the northern, yellow-winged race. When first reporting this specimen, Berlepsch (Ibis, 1884, pp. 438-439) noted that it agreed with others from southeastern Brazil, but had the bill longer, and broader at the base, and that the “. . . upper wing coverts are more yellowish, not so much mixed with brownish, and the black stripes on them thinner and finer.” The species, then, has two races, which are widely separated geo- graphically, as follows: Syrigma s. sibilatrix—southern Brazil, Uruguay, Paraguay, Bolivia, and northern Argentina. Syrigma s. fostersmuithi—north-central and northern Venezuela. At Cantaura, July 13, 1947, Mr. Smith collected an adult male short-eared owl in good plumage. No short-eared owl was previously known to occur in northern South America east of the high paramo zones of Colombia and Ecuador, and it is not surprising to find that the one collected differs sufficiently from all the described forms to warrant giving it a name. ASIO FLAMMEUS PALLIDICAUDUS, new subspecies Type—U.S.N.M. 406386, ad. 3, Cantaura, Anzoategui, Venezuela, July 13, 1947, collected by Foster D. Smith, Jr. (orig. No. 20). Subspecific characters—Resembling A. f. bogotensis Chapman of Colombia-Ecuadorian highlands in its general dark color above, but differing from that race in having the ochraceous buffy markings on the back somewhat more extensive, in having a strikingly paler tail, the central rectrices being light ochraceous buff barred broadly with fuscous, the pale and the dark bands about equal in width (the dark ones very much wider than the pale ones in bogotensis) and the lateral rectrices similar with the dark bars rapidly decreasing in width, becoming narrow bars on the inner webs of the outermost pair and entirely absent on the outer web of the outermost pair (the median rectrices in pallidicaudus are about like the outer ones in bogotensis), the facial disc more tinged with ochraceous buff, the breast less heavily streaked with fuscous, the under wing coverts almost immaculate ochraceous buff (much streaked with fuscous in bogotensis) and the outermost primaries with fewer bands on the underside, unbarred for the basal two-thirds (in bogotensis less than two-fifths). Measurements of type-—Wing 304, tail 140, culmen from base 30, tarsus 52 mm. NO. 9 HERON AND OWL FROM VENEZUELA—FRIEDMANN 3 Range.—Known only from the type locality. In his review of the neotropical short-eared owls, Bangs (Proc. New England Zool. Cl., vol. 6, pp. 95-98, 1919) rightly criticizes the tendency in literature to give the American range of the bird as North and South America from the Arctic Ocean to Patagonia, and goes on to say that he “. . . can find no record for the short-eared owl from anywhere in southern Central America, nor in the vast forested regions of northern South America. Even in migration North Amer- ican birds range only about as far as Guatemala and Cuba. In South America, except for the very distinct local form (bogotensis) .. . the short-eared owl is confined to the open regions, south of the forest, in temperate southern South America... .’ The new race described in this paper is the first evidence of a connecting link in the open areas of Venezuela. nh e ye Bee A NeuAT wert i faye fo Fuel? hae i ioe Abate ato olay Ua rete i) , vi payee BS aN gel pote ae ae Gee a hina anand & hie) a ee a ey ve wil . if Fear ee cele “yee gurantee a vie : P ss ; ah ya Fad boghgth a ie tpn wd print yi et y oO Fe hie Ee Me ek fig De is a ; ve me jeepers ts Ret out we : eh vee ' ] Rote! RE LE Oy stead el hw an % ane Cee oe | Bry ah; Blt gi oI Te Ui 4 Ey ; ‘4 1 nage SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 10 ASCOLLE GION ORI ISUEs FROM TALARA, PERU BY; SAMUEL F. HILDEBRAND AND OTIS BARTON ME-INCR lel] (PuBLicaTion 3986) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION AUGUST 18, 1949 : ' hy Te ee, vy i May xP ; y iy 1, ' Zz iM iy ity ee ie t AG i aD The Lord Baltimore Press BALTIMORE, MD., U. 8 A. A COLLECTION OF FISHES FROM TALARA, PER By SAMUEL F. HILDEBRAND! anp OTIS BARTON 2 An unusually interesting collection of fishes was obtained at Talara, Pert, from January to April, 1946, by the junior author. Two new species from this collection have already been described (Barton, 1947), and descriptions of seven others are included in these pages. In addition to the apparently new forms, three species previously not reported from Pert are included. Descriptions of these species also are offered herein, as the present paper is to serve as a supplement to “A Descriptive Catalog of the Shore Fishes of Peru” (Hilde- brand, 1946). Most of the specimens were obtained from fishermen who fished on the banks west of Talara, at depths down to 250 feet. The carefully executed drawings of the new species herein described were made by Mrs. Ann S. Green of the U. S. Fish and Wildlife Service. The paper was prepared in the Division of Fishes, United States National Museum. Family ALBULIDAE: Ladyfishes; Bonefishes Body elongate, little compressed ; head rather low, flat above ; snout conic, projecting far in advance of lower jaw; mouth moderately small, nearly horizontal; maxillary reaching nearly to front of eye or below it, never beyond eye; premaxillaries not protractile ; no gular plate; teeth all small, present in bands on jaws, vomer, palatines, pterygoids, basibranchials, and sphenoid; branchiostegals 13 or 14; gill membranes separate and free from the isthmus; gill rakers very short, stout, tubercular in large examples; lateral line straight ; scales rather small, with membranous borders, not extending on head, a modified row on back in front of dorsal fin; dorsal fin beginning in advance of ventral fins; caudal fin forked; anal fin very small, far behind dorsal; ventral and pectoral fins similar, each with an axillary scale. 1Ichthyologist, United States Fish and Wildlife Service. Dr. Hildebrand died on March 16, 1949, before this paper went to press. 2 Explorer. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III to Genus ALBULA This genus differs from Divonina, the only other genus of the family, chiefly in the last ray of the dorsal and anal being short; not produced and filamentous as in Dixonina. Other characters are in- cluded in the description of the family. ALBULA VULPES (Linnaeus) Esox vulpes LINNAEUS, 1758, ed. 10, p. 313, Bahama Islands. (Diagnosis, based on Vulpes bahamensis Catesby, pre-Linnaeus. ) Albula vulpes JorvAN and EverMANN, 1896, p. 411, fig. 179 (description; synonymy); Meek and Hildebrand, 1923, p. 179 (synonymy; description). Head 3.0, 3.4; depth 4.9, 5.0; eye 4.2, 4.0 in head; snout 3.1, 2.8; maxillary 3.0, 2.9; interorbital 5.25, 5.5; caudal peduncle 4.1, 4.0; anal base 4.5, 5.1; ventral fin 2.2, 2.3; and pectoral fin 6.2, 6.7. D. 17, 17 A. 8, 8; Ps 16; 1747scales' 70, 703 sill rakers 7-- uy oq a: Two young adults, 68 and 80 mm. in total and 55 and 64 mm. in standard length are included in the collection. The proportion or enumeration given first in each instance pertains to the larger speci- men. The family and generic characters, together with the data given in the preceding paragraph are sufficient to identify the species. These specimens retain the dark cross bands on the back of the juvenile, which generally disappear at about the length attained by the larger specimen. Range.—Reported from nearly all warm seas: On the Pacific coast of America from Monterey Bay, Calif., southward to Panama Bay, and now to northern Pert; and on the Atlantic from Woods Hole, Mass., to Rio de Janeiro, Brazil. Family MURAENIDAE: Moray eels PRIODONOPHIS EQUATORIALIS Hildebrand Priodonophis equatorialis HiLDEBRAND, 1946, p. 134, fig. 31, Cabo Blanco, Pert (original description). A single specimen, 490 mm. in total length, is included in the collec- tion. It, in general, agrees very well with the holotype and a paratype of this species. Minor differences are evident, however, from a com- parison of the specimens. The skin seems to be thinner and smoother in the Talara specimen, which may be the result of different methods of preservation, or possibly of a difference in age, as this specimen is smaller than the type and paratype. One eye in the Talara speci- men definitely is smaller (probably abnormal) than in the type speci- mens, but the other one is larger and just about bridges the gap. There seems to be virtually complete agreement in the size and shape NO. I0 FISHES FROM PERU—-HILDEBRAND AND BARTON 3 of the teeth, which are in a single series in each jaw, have broad bases, are flattened, and have definite serrations at least on the posterior margins. However, the Talara eel has two small teeth on the median line far back on the roof of the mouth, which are missing in the other specimens. In color the Talara specimen is darker brown than the other specimens, and it has fewer pale spots than the holotype, but more than the paratype, becoming larger on the distal part of the tail than in the other specimens. The differences indicated are regarded as variations within the species. Range.—Northern Pert, from off Mount Organos and from Talara. PRIODONOPHIS SERRATIDENS, new species FIGURE I Body with smooth skin, somewhat compressed, its thickness at vent about four-fifths of its depth at same place; tail more strongly com- pressed, tapering to a rather narrow point; length anterior to vent slightly greater than length posterior to vent, 1.9 in total length; head more strongly compressed than body, its width just in front of gill opening equal to about half its depth at the same place, its length anterior to gill opening 8.3 in total length and 4.25 in length anterior to vent; greatest depth (at gill opening) 6.8 in length anterior to vent, 1.6 in head; snout moderately robust, rather broader than deep in cross section, 5.1 in head; eye small, 12.2 in head, 2.4 in snout; mouth large horizontal, the gape extending about half its length be- yond middle of eye, 2.25 in head; lips with small papillae; upper lip and lower jaw with prominent pores; teeth in jaws in a single series, definitely compressed, with rather broad bases, and with both margins strongly serrated, none on vomer, though present in a single series far back on shaft (see insert, fig. 1) and none on palatines; anterior nostril with a tube about two-thirds length of eye, situated well above margin of lip, posterior nostril a round pit, situated at dorsal edge of snout, a little in advance of vertical from anterior margin of eye; gill opening an oblique slit, about 1.5 times length of eye; vertical fins very low, especially anteriorly, not incased in thick skin, fully con- fluent around the tail, the origin of dorsal a little in advance of gill opening, its origin to tip of snout 4.2 in length anterior to vent; caudal fin rather broadly rounded. General ground color brown; sides of head with dark brown longi- tudinal stripes, extending on anterior part of body, becoming cross lines and reticulations below and somewhat behind gill opening ; upper surface of head and body everywhere, except on ventral surface of head and trunk, with pale spots, mostly larger than eye, extending more or less on the dorsal and anal fins; dorsal fin with an intra- III VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS "y}99} SulMoys usdo peaids yynour ‘gq {suoIs0do1d pue adeys Surmoys weiseip ‘y *(€Szbrr ‘oN “W'N'S'Q) neg ‘elejey, ‘suo osu “wu O1$ ‘odAy 9Y} WO ‘suapyDssas Sty gouopoisg—t >> >>> ory say Vy LORE pp ‘Oly NO. IO FISHES FROM PERU-—-HILDEBRAND AND BARTON 5 marginal pale streak anteriorly, becoming marginal over the mid- caudal section, soon fading into the general pale brown color of the posterior section of the fin. This species is represented by a single specimen (U.S.N.M. No. 144253) 510 mm. in total length. Its chief distinguishing characters are shown in the parallel comparison of the three species of this genus herein recognized, which follows the description of P. angusticeps. The name serratidens is in reference to the serrated teeth in the jaws. PRIODONOPHIS ANGUSTICEPS, new species FIGURE 2 Body with wrinkled skin, compressed, its thickness at vent a little less than three-fourths its depth at same place; tail more strongly compressed, becoming strongly compressed and broadly rounded dis- tally; length anterior to vent equal to length of rest of body; head deep, rather strongly compressed, its width just in front of gill open- ing equal to about half its depth at same place, its length anterior to gill: opening 6.4 in total length and 3.25 in length anterior to vent; greatest depth (at gill opening) 6.6 in length anterior to vent, 2.0 in head; snout fairly robust, not much deeper than broad, about square in cross section, 6.1 in head; eye small, 11.4 in head, 1.85 in snout; mouth very large, horizontal, the gape extending about half its length beyond middle of eye, 2.3 in head; lips with small papillae; upper lip and lower jaw with prominent pores; teeth in jaws in a single series, rather small, not prominently compressed, with finely serrated margin visible only under magnification, none on vomer or palatines ; anterior nostril with a tube about two-thirds length of eye, situated well above margin of upper lip, posterior nostril a roundish pit sur- rounded by a slightly raised membrane, situated near edge of dorsal surface of snout just in advance of vertical from anterior margin of eye; gill opening an oblique slit, nearly twice diameter of eye; dorsal fin high, more than twice as high as the anal, its greatest height about equal to length of snout and eye, fully confluent with the caudal and anal, its origin from tip of snout 4.9 in length anterior to vent; caudal fin broadly rounded. General color rather light chocolate brown, the furrows of the wrinkled skin dark brown, mostly horizontal on head and vertical on body, some wavy or even slightly cross hatched, disappearing on distal part of tail; fins of about same color as body, dark lines following the furrows of the wrinkles in the skin, paralleling the rays. This apparently new eel is represented in the collection by a single specimen (U.S.N.M. No. 144254), 510 mm. in total length. The III VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS ‘y}90} Surmoys usdo peaids yynow ‘gq ‘suonsodosd pue odeys SurMoys wiesiseip ‘y ‘(PSzbrI ‘ON “JWN'S'N) Died ‘eseyey, “suoy “wu orf ‘adA} ay} Woy ‘soeds Mau ‘sdassnbuv siydouopolag—z “1 954 NO. IO FISHES FROM PERU-——-HILDEBRAND AND BARTON Wl principal differences among the three species of this genus, recog- nized in these pages, are shown in the parallel comparison that follows. The name angusticeps was suggested by the rather narrow com- pressed head. equatorialis Teeth in jaws com- pressed, with broad bases, the posterior margin of each tooth always and the an- terior margin gener- ally serrate. Skin on body mostly smooth, wrinkles if present not extending on dorsal fin. Head and trunk defi- nitely shorter than tail, length anterior to vent 2.3 in total length. Head large, its length to gill opening 3.0 in length anterior to vent, 6.8 in total length. Snout about square in cross section, 6.0 in head. Origin of dorsal well in advance of gill slit, its distance from tip of snout 4.1 in length an- terior to vent, 9.4 in total length. Dorsal fin rather low, its height nowhere exceeding length of snout, not enveloped in thick skin. Head and body with pale spots, very small (dots) on head, be- coming larger poste- riorly, all smaller than eye. serratidens, new species Teeth in jaws com- pressed, with broad bases, each margin of tooth definitely ser- rate. Skin on body and fins smooth. Head and trunk about equal to length of tail. Head small, its length to gill opening 4.25 in length anterior to vent, 8.3 in total length. Snout rather broader than deep in cross sec- tion, 5.1 in head. Origin of dorsal little in advance of gill slit, its distance from tip of snout 4.2 in length an- terior to vent, 8.2 in total length. Dorsal fin very low, its greatest height about half length of snout, enveloped in thin skin. Head and body with much larger pale spots, many of them equal to or larger than eye. angusticeps, new species Teeth in jaws little com- pressed, with minute serrae, present only at base of most of the teeth. Skin on body very wrinkled, the wrinkles extending on dorsal fin. Head and trunk about equal to length of tail. Head moderately large, its length to gill open- ing 3.25 in length an- terior to vent, 6.4 in total length. Snout a little deeper than broad in cross section, 6.1 in head. Origin of dorsal far in advance of gill slit, its distance from tip of snout 4.9 in length an- terior to vent, 9.6 in total length. Dorsal fin high, its greatest height about equal to length of snout and eye. Head and body un- spotted, but with dark lines in the furrows of the wrinkled skin. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Family GADIDAE: Codfishes Body generally quite elongate, tapering toward the tail; mouth large, terminal or more usually inferior; chin with a barbel; gill openings wide ; gill membranes separate or somewhat united, generally free from the isthmus; gills 4, a slit behind the fourth; vent generally more or less median in position ; air bladder rarely missing in adults ; scales small, cycloid ; fins without spines; dorsal fin generally occupy- ing nearly the full length of back, single or divided into 2 or 3 sections ; caudal fin separate or united with the dorsal and anal; anal fin long, single or divided into 2 parts; ventral fins jugular, each consisting of 1 to 8 rays. This is a large family, which includes many important food fishes. In tropical regions it is represented by species inhabiting deep water. Genus PHYSICULUS Kaup, 1858 Body elongate, robust anteriorly, tapering sharply posteriorly ; head large, generally more or less depressed; mouth broad, more or less inferior; snout broad; chin with a short barbel; teeth in jaws in villiform bands, none on vomer or palatines; scales small, extending forward on snout and on chin; dorsal fins 2, the first one small; caudal fin round, free from dorsal and anal; anal fin single; ventral with 5 to 7 rays, the outer ones more or less filamentous. About seven species are known from off the American coasts, generally living in rather deep water. PHYSICULUS TALARAE, new species FIGURE 3 Head 3.85; depth 4.6; D. 10-60; A. 63; P. 26 and 27; scales partly lost, about I10. Body robust anteriorly, tapering sharply and becoming rather strongly compressed posteriorly, its depth at base of pectorals scarcely an eye’s diameter greater that its width at the same place; caudal peduncle very slender, strongly compressed, its depth 11.5 in head; head rather large, a little broader than deep at margin of preopercle, its depth at this point 5.5 in standard length; snout low and broad, 4.1 in head; eye moderate, 4.4; interorbital flat, 4.75; mouth rather large, its gape about as broad as long; lower jaw definitely shorter than the upper one, included ; maxillary almost reaching vertical from posterior margin of eye, 1.9 in head; teeth in jaws in villiform bands, NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON aM ELS tear goo te Wh Fic. 3.—Physiculus talarae, new species. From the type, 265 mm. long, Talara, Peru (U.S.N.M. No. 144255). {0 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL satiate the one in upper jaw the broader, mostly exposed with mouth closed ; preorbital scarcely as broad as pupil; gill rakers short, expanded at tips, spiny, 12 on lower limb and 4 on the upper one of first arch; lateral line incomplete, ending near midbody length, or about 2 diameters of eye behind tip of pectoral, with a long low arch an- teriorly, being highest under origin of second dorsal; scales small, extending forward to margin of snout and on chin, but not on fins except on base of caudal, 8 longitudinal rows between lateral line and middle of first dorsal; dorsal fins 2, definitely separate, the first short, more or less triangular in shape, its middle rays longest, 2.33 in head, the second long and low, with a straight margin; caudal fin small, round, about as long as snout and half the eye; anal fin similar to second dorsal; ventral fin narrow, with 6 rays, the outer rays produced, the second the longest, reaching about an eye’s diameter beyond origin of anal, but failing to reach tip of pectoral by about an equal distance, definitely shorter than head, 5.3 in standard length ; pectoral rather large, pointed, shorter than head, 5.3 in standard length. Color uniform gray, except for underneath surface in advance of origin of anal which is darker brown; margin of gill covers quite dark; dorsal pale, with a brownish margin; caudal rather darker brown than body, anal similar to dorsal, except that it has a darker margin; ventral dark at base, otherwise colorless; pectoral brownish like the body. This apparently new species is represented by a single specimen (U.S.N.M. No. 144255), 265 mm. in total and 243 mm. in standard length. It seems to be nearest P. nematopus Gilbert, known from many specimens taken at depths ranging from 71 to 221 fathoms in the Gulf of California, of which five “type” specimens, 78 to 147 mm. in total length (U.S.N.M. No. 46555), are at hand for comparison. The most outstanding difference between those specimens and the one from Pert is the difference in the number of pectoral rays. This and other differences are shown in the parallel comparison that follows. Some of the other differences, consisting of proportions such as the depth of the body and the length of the ventral fins, may be affected by age and growth, and therefore are not entirely reliable. Another closely related species is P. longipes, known from speci- mens taken in 127 to 695 fathoms in Panama Bay or in the vicinity thereof, of which one of the “type” specimens, 113 mm. in total length (U.S.N.M. No. 57876), is at hand. The Peruvian specimen differs from this one in the greater number of pectoral rays, just as NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON II it does from nematopus, but it agrees in the depth of the body and more or less in the length of the head. These and other differences are shown in the parallel comparison. The validity of this species, that is, its distinctness from nematopus, has been questioned by Norman (1937, p. 56). In view of the present study it seems ad- visable to retain nematopus and longipes as distinct species. Garman (1899, p. 189) stated that longipes differs from nematopus in having a greater number of dorsal rays and a smaller number of ventral rays. All specimens of 4 species examined have uniformly 6 ventral rays, and the difference in the number of dorsal rays is at most very small, and instead of being more numerous, they actually are slightly fewer according to the specimens examined. The only other species of this genus reported from off the Pacific coast of tropical America, rastrelliger, of which several “type” speci- mens are at hand (U.S.N.M. No. 44281), differs from all the others of that region in having more gill rakers, 18 to 20 being present on the lower limb of the first arch. nematopus Pectoral rays 20 to 23. Dorsal rays 10-56 to 50. Scales in lateral series 86 to about 100, and 6 longitudinal rows between lateral line and middle of first dorsal. ‘ Ventral fin longer than head, 3.0 to 3.55 in standard length. Head moderate, its length 3.9 to 4.0 in standard length. Depth 5.1 to 5.25 in standard length. Eye 3.4 to 4.1 in head. Gill rakers blunt, not definitely expanded at tips, quite spiny, nearly as long as pupil. longipes Pectoral rays 20 and 21. Dorsal rays II-54. Scales in lateral series lost in part, about 76, and 6 longitudinal rows between lateral line and middle of first dorsal. Ventral fin longer than head, 3.3 in standard length. Head large, its length 3.5 in standard length. Depth 4.7 in standard length. Eye 4.65 in head. Gill rakers pointed, not expanded at tips, not very spiny, nearly as long as pupil. talarae, new species Pectoral rays 26 and 27. Dorsal rays 10-60. Scales in lateral series lost in part, about 110, and 8 longitudinal rows between lateral line and middle of first dorsal. Ventral fin shorter than head, 5.3 in standard length. Head moderate, its length 3.8 in standard length. Depth 4.6 in standard length. Eye 4.4 in head. Gill rakers broad, greatly expanded at tips, very spiny, scarcely half length of pupil. 12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT Family SERRANIDAE: Seabasses Genus ANTHIAS Bloch, 1792 Body oblong, compressed ; mouth large; premaxillaries protractile ; maxillary exposed, covered with scales, with a supplemental bone; teeth in jaws villiform, intermixed with canines, a triangular patch on vomer and an elongate one on palatines, a few or none on tongue; gill rakers long and slender; preopercle serrate, without antrorse spines ; lateral line complete, running close to dorsal outline, abruptly decurved under last rays of dorsal; scales covering head and body, smooth or ctenoid ; dorsal fin X, 12 to 18; anal fin III, 6 to 8; pectoral fin about 17 or 18; ventral fin long, inserted under base of pec- » toral, I, 5. This genus was not represented in the collections on which U. S. National Museum Bulletin 189, 1946, was based. However, Mr. - Barton obtained two specimens from fishermen at Talara, Pert, from January to April, 1946, which he described (1947, p. 2) under the name Holanthias sechurae. It seems now, however, that the nominal genus Holanthias is not well founded, and that it is advisable to refer the species to the genus Anthias for the present, along with other related American species. This genus, although widely distributed in tropical and temperate seas, until recently was known from the Western Hemisphere from only one species reported from the Atlantic coast of South America. Two very closely related species were described recently from the eastern Pacific, one from off Cape San Lucas, México, and the other from Talara, Pert. ANTHIAS SECHURAE (Barton) FIGURE 4 Holanthias sechurae Barton, 1947, p. 2, fig. 2, Talara, Peri (description, based on the holotype, 235 mm. in total length (A.M.N.H. No. 17082), and a paratype 192 mm. in total length). Head 3.1; depth 3.1; D. X, 15; A. III, 7; P. 10; scales 55moceea, 52 or 53 in lateral line. Body rather deep, compressed, dorsal outline more strongly convex than the ventral; snout a little longer than eye; eye 4.0 in head measured to tip of opercular spine; interorbital 4.0; mouth oblique; lower jaw projecting; maxillary extending just beyond center of eye, its greatest width two-thirds diameter of eye; tongue with a considerable patch of fine granular teeth; upper jaw anteriorly with small canine teeth; lower jaw with similar teeth anteriorly and lat- 13 7 FISHES FROM PERU-——-HILDEBRAND AND BARTON Io NO. *(ZQ0Z1 (‘ZP61 ‘uoyIeg J91;v) ‘ON (H'N'W'YV) aed ‘erejey ‘suo, ‘uu Sez % 05: of PONG 8 4, Deg MUle aE Nh grte at ee an wis a ON be NaN Saye RI an ‘adA} 94} Wolly ‘(UOJIeG) avD4ényIas so1yJup—TP “DI Mia, is EI) es aaste y J *} ray) YA abs bes 0% ee if 7 #9185 ie MEK OOM Ath gt 14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III erally, the anterior ones projecting obliquely forward; preopercle finely serrate, the paratype with two small flat spines at angle in addi- tion to fine serrations on vertical margin; opercle with three spines, the middle one the largest; gill rakers long, slender, 26 on lower limb of first arch; lateral line running close to back, abruptly decurved under last rays of dorsal, thence a little below middle of caudal peduncle to base of caudal fin; scales present on head and body, ex- clusive of premaxillary, and upper border and tip of mandible; dorsal fin beginning above middle of opercle, the spines increasing in length to the third, then slowly decreasing, the third spine 2.3 in head; soft dorsal with gently convex margin, middle rays longest, no notch be- tween spinous and soft portions; caudal forked, the outer rays not produced; anal fin small, its origin a little behind vertical from be- ginning of soft part of dorsal, its first spine short, the second one nearly as long as the third, and stronger; ventral fin with first and second soft rays filamentous, the second the longer, extending about to end of anal base; pectoral with narrowly rounded margin, extend- ing a little beyond origin of anal, 1.2 in head. Color of type “‘masked by purple stain.’’ Paratype, “yellowish with brown mottling above, a little paler below, fins all pale. It shows traces of three radiating dark lines behind the eye, each line about as wide as pupil, a dark patch between the eyes extending onto snout, and a dark line in front of the eye carried onto tip of lower jaw.” (Reorganized and largely reworded after Barton’s description and figure. ) This species was described by Mr. Barton, as shown above, from two specimens, the type 235 mm. in total and 188 mm. in standard length, and a paratype 192 mm. in total length, both from the vicinity of Talara (exact place of collection uncertain), Pert. It already has been pointed out by John T. Nichols in a footnote in Mr. Barton’s paper that this species is close to Anthias gordensis Wade (1946, p. 225). A. gordensis was described from two small specimens, 127 and 129 mm. in standard length, taken at 70 to 78 fathoms on Inner Gorda Banks, off Cape San Lucas, México. Whether sechurae and gordensis actually are distinct cannot be definitely determined from the published accounts. The number of fin rays and gill rakers, indeed, are virtually the same in the two, but a slight difference in the number of scales in a lateral series is in- dicated, 55 or 56 being given for sechurae and 48 to 50 for gordensis. It is not stated, however, exactly where and how the enumerations were made. The eye may be a little smaller in sechurae, 4.0 in head, than in gordensis, 3.61 and 3.75 in head, but then the specimens of NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON 15 sechurae are larger. The snout was described as a little longer than the eye in sechurae and as shorter than the eye in gordensis. The teeth on the tongue were described as granular and as in a “‘consider- able patch” in sechurae, and as blunt, few and scattered in gordensis. According to the figures the lobes of the caudal are pointed in sechurae and about as long as the head without the snout, whereas they are rounded and almost as long as the head in gordensis. Although the number of anal rays was given as equal in the two species, the illus- trations show the base to be shorter than the soft dorsal by 5 rays in sechurae, and by only 2 rays in gordensis. Furthermore, the mouth is shown as notably more oblique in sechurae than in gordensis. In view of these several apparent differences the two may be regarded as distinct at least until further evidence is obtained. This species, as well as A. gordensis, differs from A. asperilinguis from the Atlantic coast of South America in having a more elongate body, smaller scales, longer snout, smaller eye, and in the absence of filaments on the outer rays of the caudal. Range.—Known only from the vicinity of Talara, Pert. Family PRIACANTHIDAE: Bigeyes Body oblong, or ovate, compressed ; head short, deep; snout short; eye very large; mouth large, oblique; lower jaw projecting; teeth in jaws in villiform bands, present also on vomer and palatines; pos- terior nostril large, elongate ; preopercle serrate, with I or more spines or enlarged serrations at angle; opercle short, with 1 to 3 points or spines ; lateral line complete, not extending on caudal fin; scales firm, ctenoid, extending on head, snout and maxillaries, but not on the fins; dorsal fin continuous, with about ro spines; anal with 3 spines; ventrals thoracic, with I, 5 rays. Two genera, Priacanthus and Pseudopriacanthus, occur in Ameri- can waters, which have not heretofore been recorded from Pert. The last mentioned genus is represented by a fine specimen in the collection from Talara, Pert. Genus PSEUDOPRIACANTHUS Bleeker, 1869 Body very deep, its depth usually equal to or greater than half its length to the base of caudal; scales moderately large, about 35 to 55 in a lateral series; dorsal with about X, Io or I1 rays; anal with about LI], 10or 11. This genus, which occurs in the Atlantic and Pacific, is new to the fauna of Pert. 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III PSEUDOPRIACANTHUS SERRULA (Gilbert) Priacanthus serrula GILBERT, 1890, p. 450, Albatross Station 2797, 8°6’30” N., 78°51’ W., Panama Bay, in 33 fathoms (original description). Pseudopriacanthus serrula JorDAN and EverMANN, 1896, p. 1239 (description) ; Meek and Hildebrand, 1925, p. 490 (description). ?Pseudopriacanthus lucasanus CLARK, 1936, p. 388, Cape San Lucas, Baja California (original description). Head 2.6; depth 1.8; D. X, 11; A. III, 10; P. 17; seales 53. Body short and deep, moderately compressed, the ventral outline anteriorly more strongly convex than the dorsal; caudal peduncle short and deep, 2.8 in head; snout only about half length of eye, 4.35 in head; eye extremely large, 2.25; interorbital 4.3; mouth strongly oblique; lower jaw projecting prominently, entering general dorsal outline of head; maxillary nearly as broad as pupil, not quite reaching vertical from anterior margin of pupil, 1.85 in head; teeth in jaws in villiform bands, some of the outer ones in anterior part of each jaw slightly enlarged, also present in villiform bands on vomer and palatines; preorbital bone scarcely half width of pupil, rather finely serrate above and below; supraorbital ridge finely serrate; preopercle with fine serrae on its vertical margin and with larger ones at its angle and on the horizontal margin; gill rakers fairly short, 17 on lower limb of first arch; lateral line arched anteriorly, running rather close to back; scales small, strongly ctenoid, extend- ing forward on head, covering it fully except for the premaxillaries and the lower lips; dorsal fin long continuous, the spines fluted, graduated to fourth, the fourth to seventh of nearly equal length, the rest shorter, the fifth 1.45 in head, the margin of the soft part convex, the rays spinous at base, the longest ones only a little shorter than the longest spines; caudal fin round, about as long as head without snout; anal with three graduated, fluted spines, the third 2.3 in head, the longest soft rays much longer than the longest spine, about equal to longest rays of dorsal; ventral fin large, reaching opposite base of second anal spine, nearly as long as head, its spine 1.1 in head; pectoral fin shorter, rather broadly rounded, 1.6 in head. Color uniform light gray, scarcely paler below than above; fins uniform pale except for dark tips or margins of the soft parts of the dorsal and anal, and the caudal and ventral. The description was based on a single large specimen, 280 mm. in total length, which was taken off Talara, Peru. It was compared with the small type, which is only 38 mm. in total length. The comparison, although not entirely satisfactory, shows fair agreement. The two agree in the number of fin rays present, and perhaps more signifi- NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON 17 cantly the two agree in the small number of gill rakers present, wherein this species seems to differ from related ones. It is close to P. altus, from the West Indies and the Atlantic coast of the United States, apparently differing principally in having 17 gill rakers on the lower limb of the first arch, instead of 20 asin P. altus. P. serrula also is close to P. niphonius, a Japanese species, from which it also differs in having fewer gill rakers, a broader interorbital, and rather longer fin spines. The small specimen, 73 mm. in total length, from Cape San Lucas, Baja California, described by Clark (1936, p. 388) as P. lucasanus, probably is the same as P. serrula, though a few characters as given are in disagreement. The anal formula, “II, 10,” perhaps may be dismissed as a typographical error, as the presence of 3 spines is the normal number for the members of the family. The oversight of teeth on the vomer and palatines, also a family char- acter, presumably led to the assertion than none were present. Range.—Previously reported only from Panama Bay. The known range is now extended southward to Talara, Pert, and somewhat doubtfully, northward to Cape San Lucas, Baja California. Family POMADASIDAE: Grunts ORTHOPRISTIS CHALCEUS (Giinther) Orthopristis chalceus HILDEBRAND, 1946, p. 284 (description). A single specimen, 59 mm. in standard length (caudal fin broken), is present in the collection. The species probably is not common in Pert, as the U. S. Fish and Wildlife Service Mission to Pert in 1941 did not obtain any specimens. It has been recorded, however, from two places in Pert, namely, from Lobos de Afuera, and from Callao (Hildebrand, 1946, p. 248). Range.—Gulf of California to the Galapagos Islands and northern Peru Family SCIAENIDAE: Croakers, Drums, etc. Genus EQUETUS Rafinesque, 1815 Body oblong, compressed ; back much elevated anteriorly, descend- ing rapidly posterior to first dorsal fin; mouth small, inferior, lower jaw included ; snout with rather prominent pores and slits; preopercle with serrated membranous border; teeth in jaws in villiform bands, some of them occasionally enlarged; gill rakers short and rather few, about 8 to 12 on lower limb of first arch; scales rather small, ctenoid ; soft part of dorsal very long, with about 35 to 55 rays; anal small, with only about 5 to 8 soft rays. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT This genus was not known from Pert when U.S. National Museum Bulletin 189, 1946, was prepared. However, Mr. Barton obtained a specimen of this genus at Talara, Pert, in 1946, which he described (1947, p. 1) as Eques * lanfeart. EQUETUS LANFEARI (Barton) FIGURE 5 Eques lanfeari Barron, 1947, p. 1, fig. 1, Talara, Peru, from a depth of 250 feet. (Description, based on the holotype (A.M.N.H. No. 17081), 290 mm. in total length.) Head 3.2; depth 2.7; D. XII, 35 or 36; A. I, 7; scales in 80 trans- verse series above lateral line, 12 rows between lateral line and middle of first dorsal. Body deep, much compressed ; head compressed ; snout blunt, not protruding beyond premaxillaries, 3.0 in head; eye 4.7; interorbital 3.7; mouth nearly horizontal; maxillary reaching about under middle of eye, 2.6 in head; tip of lower jaw with a fleshy knob; posterior nostril oval, the anterior one somewhat triangular and smaller; teeth in wide bands in both jaws, the outer ones in upper jaw enlarged; gill rakers very short, 9 exclusive of rudiments on lower limb of first arch; lateral line not distinct, rather strongly arched; scales strongly ctenoid, extending on soft dorsal, caudal, and anal; dorsal fins barely continuous, spinous dorsal short, the first spine very short, the second and third high, the third 1.8 in head; second dorsal very long and rather low ; caudal fin rounded; anal small, the second spine enlarged, 2.2 in head; ventral nearly as large as the pectoral, inserted almost under base of pectoral, 1.6 in head; pectoral 1.7 in head. Color gray, purplish along back, a little lighter on belly; all fins dusky at edges; snout, opercles and preopercles darker; four distinct, horizontal, dark stripes, the first about one-third of the pupil in width, starting below middle of first dorsal and running along base of fin to middle of soft dorsal; the second, about one-half of the pupil in width, starting about the length of snout below the first dorsal spine and curving backward to base of third hindermost ray of soft dorsal ; the third slightly wider, from back of opercular angle to upper third of peduncular base; the fourth a little narrower, starting a little be- hind lower pectoral base and running back to above posterior anal base. (Reorganized and somewhat reworded after Barton.) 8 Eques Linnaeus (1758, p. 459) is preoccupied in Lepidoptera, but Equetus Rafinesque (1815, p. 86) is available. 7 I PERU-—HILDEBRAND AND BARTON 19 FISHES FROM Io NO. (‘2r61 ‘uoyeg Joipy) *(1g041 ‘ON "H'N'IW'Y) niog ‘ervey, ‘suo, ‘wut 062 ‘adAq oy} Wor (uojreg) tdvafun] snjanby—sS -o1g 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III This species probably is nearest E. acuminatus (Bloch and Schneider) from the Atlantic, from which it differs prominently in the much smaller scales. From the other two species, E. viola (Gilbert) and £. fuscovittatus (Kendall and Radcliffe), known from the Pacific coast of America, it differs in color and in the lower spinous dorsal. Range.—Known only from off Talara, Peru. Family CHAETODONTIDAE: Butterflyfishes, Angelfishes Genus HOLACANTHUS Lecépéde, 1803 Body oblong, rather robust, the depth usually about equal to half the total length; back round; head short, blunt; preopercle serrate along vertical margin, with a large spine at angle, preceded by 1 to 3 spines in lower margin of preopercle; interopercle anteriorly with I to 4 short spines; preorbital anteriorly with coarse serrae; scales firm, strongly serrate, fluted, accessory scales generally present; dorsal with about 12 to 15 spines; soft part anteriorly sometimes elevated; anal with three graduated spines, the soft part similar to that of dorsal. Color usually brilliant. The genus is now reported from Pert for the first time. HOLACANTHUS PASSER Valenciennes Holacanthus passer VALENCIENNES, 1855, p. 327, Galapagos Islands (original description; figure published in atlas, 1846, as plate 6) ; Jordan and Ever- mann, 1898, p. 1682 (description); Meek and Hildebrand, 1928, p. 778 (description; range). Holacanthus strigatus GiLt, 1862, p. 243, Cape San Lucas, Baja California (original description). Head 4.0; depth 1.65; D. XIV, 19; A. III, 19; P. 17; scales along middle of side 46. Body short and deep, well compressed, its ventral profile anteriorly strongly convex, its dorsal profile slightly concave over eyes, steep at nape ; caudal peduncle short, strongly compressed, 1.6 in head ; head short, deep; snout blunt, 2.6; eye small, 4.5; mouth small, terminal ; maxillary under preorbital, reaching about to vertical from anterior nostril, 3.3 in head; teeth in jaws, slender, bristlelike, movable, mostly in 2 series; preopercle with a rather finely serrate vertical margin, at angle, a large spine exceeding diameter of eye, preceded on lower margin by 3 smaller spines; interopercle anteriorly with 2 spines; preorbital anteriorly coarsely serrate; gill rakers short, 12 somewhat developed on lower limb of first arch; lateral line pores undeveloped ; scales strongly ctenoid, fluted, mostly with accessory scales at base, extending forward on snout and chin, covering soft parts of the NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON 21 vertical fins completely, and extending more or less on the rays of the paired fins; dorsal fin very long, its origin a little in advance of margin of opercle, its spines rather short, slightly graduated, the third 2.3 in head, the soft part with a produced lobe about twice as long as head; caudal round, about as long as head; anal with 3 graduated spines, the soft part similar to that of dorsal; ventral fin long, the 2 outer rays filamentous, the longest filament reaching origin of anal, about 3.0 in length, its spine slender, 1.45 in head; pectoral broad, with slightly convex margin, about as long as head. Color satin-black; a white vertical bar under sixth dorsal spine, about as broad as eye, ending under tips of middle rays of pectoral ; caudal, ventrals and pectorals and distal parts of lobes of dorsal and anal pale; the margin of the caudal black. This species, which is new to the fauna of Pert, is represented in the collection by one fine specimen, 220 mm. in total length. Range.—Previously known from Acapulco, México, to the Gala- pagos Islands. Now for the first time reported from Pert from a specimen caught off Talara, Peru. Family CALLIONYMIDAE: Dragonets Genus SYNCHIROPUS Gill, 1860 The single small specimen in the collection, which apparently rep- resents a new species, seems to belong to the genus Synchiropus Gill, as understood by at least some modern ichthyologists.* This genus seems to be characterized by the broad, smooth body ; the absence of a tentacle above the eye; the single lateral line; the superior gill open- ing, with at least a slight free opercular flap in front of and below it; and by the two well-separated dorsal fins, with the rays in the second fin divided in adult fish. SYNCHIROPUS TALARAE, new species FIcureE 6 Head 3.1; depth 9.3; D. IV, 9; A. 8; P. 24; V. 6. Body strongly depressed, broader than deep except at base of caudal ; head large, very strongly depressed, its depth only about half its width, and 3.0 in its length; snout depressed, triangular, much shorter than eye,-4.7 in head ; eye definitely more superior than lateral, 3.5; interorbital very narrow, slightly grooved, about 12 in eye; 4I am indebted to Dr. Leonard P. Schultz, who has recently examined many specimens of the family Callionymidae from many parts of the world to determine generic and specific relationships, for suggesting that this Peruvian species belongs to the genus Synchiropus. (S.F.H.) (See Schultz and Woods, Journ. Washington Acad. Sci., vol. 38, No. 12, pp. 419-420, 1948.) Tit VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS “(BSzPrl ON ‘WN'S'Q) Deg ‘ereyey ‘Suoy ‘wut SE ‘adA} oy wos ‘satoads Mou ‘apsnjp} sndomyauksS—9 ‘OI > = oe ae: eT SE NOS LO FISHES FROM PERU—-HILDEBRAND AND BARTON 23 mouth rather large, lower jaw shorter than upper, included ; maxillary reaching nearly to anterior margin of pupil, 3.75 in head; teeth in jaws minute, apparently in a narrow band; preopercular spine strong, curved upward posteriorly, bifurcate distally, the upper spine the stronger, directed upward; gill opening superior, being an elongate tranverse slit ; lateral line single; first dorsal with four slender spines, each bearing a short filament, none reaching second dorsal without filament, origin of fin over the space between gill opening and base of pectoral, distance from snout 3.1 in standard length; second dorsal well separated from the first, the anterior rays a little higher than the posterior ones, each with a short filament and all falling far short of reaching base of caudal with filaments included; caudal fin appar- ently somewhat rounded about equal to distance from tip of snout to base of preopercular spine ; anal with widely separated rays, increasing gradually in length, but not reaching base of caudal, each ray with a short filament, origin of fin a little nearer base of caudal than tip of snout, its distance from tip of snout 1.8 in standard length; ventral rather large, inferior, the rays slightly filamentous distally, longest filament scarcely reaching opposite vent, inserted nearly equidistant from tip of snout and vent, 1.3 in head; pectoral inserted about over midlength of ventral, the rays filamentous distally, reaching slightly beyond origin of anal, 1.8 in head. General color brownish above, pale underneath; back with a dark cross line at base of first ray of second dorsal and another one at base of its last ray; side with indefinite dark spots and punctulations ; first dorsal with a black spot at base of last ray; rays of second dorsal with an elongate dusky spot somewhat above midlength ; caudal fin with a dark bar on its base, its midsection and its margin slightly darker than rest of fin; each anal ray, exclusive of the first two, dis- tally dusky, the filaments pale; ventral pale, with a few dusky dots along the rays; pectoral with a dusky spot at base of upper rays, fin otherwise plain translucent. This apparently new species is represented in the collection by a single specimen, the holotype (U.S.N.M. No. 144258), 35 mm. in total and 28 mm. in standard length, which is in good condition except that the abdomen has been excavated. S. talarae is the third species of this genus to be described from the Pacific coast of America. The other species are S. atrilabiatus (Garman) (1899, p. 122), described from specimens dredged by the Albatross off Colombia in 112 fathoms and in Panama Bay in 127 fathoms; and S. garthi (Seale) (1940, p. 36, pl. 3), described from a specimen taken at Port Utria, northern Colombia, by one of the Allan Hancock expeditions. The first-men- tioned species is well described, but the size of the specimens upon 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III which it is based is not stated and no figure is offered. The species mentioned last is rather inadequately described, but it was figured. The length of the single specimen upon which the description and figure were based was given as 30 mm. The differences among the three species as determined from descriptions and in part from the figure of S. garthi are set forth in the parallel comparison offered. S. talarae seems to be unique in the very low broad head. atrilabiatus Dorsal rays IV-o. Pectoral rays 23. Depth of body ? in head. Depth of head two- thirds its width; its width scarcely two- thirds its length. Length anterior to vent less than half the length to base of cau- dal. Snout as long as eye. Eye “less than one-third of length of head.” Gill opening ? Posterior rays of dor- sal and anal reaching base of caudal. Anal and ventral rays with filaments (“fringed”). Upper part of body with numerous closely placed brown cross streaks; first dorsal with a dark spot on outer half between the third and fourth rays. garth Dorsal rays IV-8 (ac- cording to figure). Pectoral rays 15 (ac- cording to figure). Depth of body 2.0 in head. Depth of head ? in its width; its width ? of its length. Length anterior to vent about one-half total length (according to figure). Snout one-half “width of eye.” Eye 2.5 in head. Gill opening round. Posterior rays of dorsal and anal not nearly reaching base of cau- day (according to fig- ure). None of fin rays with filaments (according to figure). Body without cross streaks, but with many brown specks; first dorsal dusky at base, but without a black spot. talarae, new species Dorsal rays IV-o. Pectoral’ rays 24. Depth of body 3.0 in head. Depth of head one-half its width; its width a little less than half its length. Length anterior to vent a little more than half the length to base of caudal. Snout about three- fourths length of eye. Eye 3.5 in head. Gill opening elongate. Posterior rays of dorsal and anal not nearly reaching base of cau- dal. Nearly all rays of fins with short filaments. Body with a very nar- row brownish cross streak at base of first ray of second dorsal and another at base of its last ray, and with indefinite dark spots and punctula- tions; first dorsal with a large black spot. NO. I0 FISHES FROM PERU—-HILDEBRAND AND BARTON 25 Family BROTULIDAE: Brotulid eels BROTULA ORDWAY], new species FIGURE 7 Head 4.25; depth 4.4; D. 110; A. 85; P. 23; scales about 225 (too small and irregular to enumerate accurately). Body moderately elongate, rather strongly compressed throughout, its thickness at tips of pectoral fins only about half its depth at same place; head compressed, rather narrow and definitely convex above, its depth at margin of preopercle 6.0 and its width at same place 9.1 in standard length; snout a little longer than eye, 4.1 in head; eye elongate, 5.4; interorbital narrow, convex, 8.0 in head; mouth mod- erate, slightly oblique; lower jaw notably shorter than the upper, included ; maxillary broad, reaching vertical from posterior rim of orbit, 2.25 in head; anterior pair of barbels of snout larger than the lateral pair and of about the same length as the nasal barbels, 2 prom- inent flaps of skin between the anterior and lateral barbels, the 6 mandibular barbels all of about equal length; a pair of pores between the anterior barbels, and 1 between the anterior and the lateral barbels; lips rather thick, rugose; teeth small, pointed, in bands on jaws, vomer, and palatines, the band on vomer broadly triangular, approaching a U-shape; 3 gill rakers and 10 rudiments on lower limb, and 4 rudiments on upper limb of first arch; lateral line running high, broadly arched anteriorly, nearly complete ; scales small, strongly striate, extending forward on the snout, and on the fins, about 30 oblique rows between posterior rim of orbit and base of opercular spine ; dorsal fin long and low, its origin about half an eye’s diameter in advance of base of pectoral, its distance from tip of snout 4.0 in standard length, the longest rays about as long as snout and half the eye; caudal fin rounded, fully united with the dorsal and anal, only a little longer than snout; anal similar to the dorsal, its origin well in advance of midlength without caudal fin, its distance from tip of mandible 2.4 in standard length; ventrals short, the longest filament 3.1 in head; pectoral fin broadly rounded, scarcely reaching halfway to origin of anal, 2.0 in head, 8.7 in standard length. Color very dark brown; sides and upper surface of head and body, to or a little beyond vertical from origin of anal, with many round black spots, the largest ones nearly as big as pupil; ventral fins dusky ; other fins very dark brown, nearly black, with a very narrow pale margin. The foregoing description is based on the holotype (U.S.N.M. No. 144259) and only specimen known, which is 375 mm. in total and 347 mm. in standard length. It differs rather prominently from the SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES Tir Fic. 7.—Brotula ordwayi, new species. From the type, 375 mm. long, Talara, Pertti (U.S.N.M. No. 144259). NO. 10 FISHES FROM PERU—-HILDEBRAND AND BARTON 27 other American species recognized by Hubbs in his revision (1944, pp. 162-178), as shown in the parallel comparison that follows. The species was named for Samuel Ordway, the judicious trustee of the New York Zoological Society. barbata Body quite elongate, its depth 5°5 to 6.0 in standard length. Head low, somewhat flattened above, its depth at margin of preopercle 6.6 to 6.9 in standard length. Mouth large, the maxil- lary reaching well be- yond posterior rim of orbit; 2/0) to *2.r im head; jaws of about equal length. Vomerine teeth in a tri- angular patch, each arm of triangle with a constriction, nearly but not quite pinching off a posterior patch. Origin of dorsal behind base of pectoral, its distance from tip of snout 3.6 to 3.8 in standard length. Origin of anal at mid- length without caudal fin, its distance from tip of lower jaw 1.9 in standard length. Ventral fins rather long, the longest filament reaching a little be- yond base of pectoral, 1.8 to 2.2 in head. Pectoral fin reaching much less than half- way to origin of anal, 8.4 to 8.8 in standard length, with 24 or 25 rays. clarkae Body quite elongate, its depth 5.9 in standard length. Head low, somewhat flattened above, its depth at margin of preopercle 6.7 in stand- ard length. Mouth large, the maxil- lary reaching rather far beyond eye, 2.0 in head; lower jaw slightly longer than the upper. Vomerine teeth roughly in a triangular patch, rounded at apex, arms of triangle without a constriction. Origin of dorsal behind base of pectoral, its distance from tip of snout 3.6 in standard length. Origin of anal a little in advance of midlength without caudal fin, its distance from tip of lower jaw 2.1 in stand- ard length. Ventral fins long, reach- ing about an eye’s di- ameter beyond base of pectoral, 2.1 in head. Pectoral fin reaching about halfway to ori- gin of anal, 83 in standard length, with 27 rays. ordwayi, new species Body deeper, its depth 4.4 in standard length. Head more convex above, its depth at margin of preopercle 6.0 in standard length. Mouth smaller, the max- illary reaching under posterior rim of orbit, 2.25 in head; lower jaw notably shorter than upper, included. Vomerine teeth in a broad triangular patch, approaching the shape of a U, the arms with- out a constriction. Origin of dorsal defi- nitely in advance of base of pectoral, its distance from tip of snout 4.0 in standard length. Origin of anal rather far in advance of mid- length without caudal fin, its distance from tip of lower jaw 2.4 in standard length. Ventral fins short, the longest filament scarcely reaching base of pectoral, 3.1 in head. Pectoral fin scarcely reaching halfway to origin of anal, 8.7 in standard length, with 23 rays. barbata Scales moderately small, about 210 in a lateral series, about 32 be- tween posterior rim of orbit and base of oper- cular spine. Head and body (at least in adults) plain, with- out round black spots ; the dorsal and anal fins more or less pale at base, with dark SMITHSONIAN MISCELLANEOUS COLLECTIONS clarkae Scales very small, about 220 in a lateral series, about 45 between pos- terior rim of orbit and base of opercular spine. Head and body (at least in adults) plain, with- out round black spots; the dorsal and anal pale at base, with dark margins. VOL. III ordwayi, new species Scales very small, about 225 in a lateral series, about 30 between pos- terior rim of orbit and base of opercular spine. Head and anterior part of body profusely spotted with black; the dorsal and anal black, with narrow pale margins. margins. Family OPHIDIIDAE: Cusk eels Genus LEPOPHIDIUM Gill, 1895 Body quite elongate, compressed, tapering regularly and rather gradually to more or less of a point posteriorly ; head low, somewhat compressed ; snout with or without a spine (present in all Pacific coast species) ; mouth rather large and broad, nearly horizontal; teeth in jaws in villiform bands, the outer ones in each jaw more or less enlarged, rather shorter and blunter on vomer and palatines; gill rakers short and few; lateral line present at least anteriorly; scales very small, with many radiating striae, extending forward on head to or beyond interorbital region ; dorsal and anal fins continuous with the caudal fin. Occurring in rather deep water on both coasts of America. About nine species have been recognized. Now reported from Pert for the first time. LEPOPHIDIUM NEGROPINNA, new species Ficure 8 Head 4.6; depth 7.5; D. 128;5 A. 114;5 P. 23;5 scales 220.5 Body moderately slender, compressed, its greatest thickness about two-thirds its depth; head long, low, a little deeper than broad over margin of preopercle; snout blunt, 4.8 in head, with a strong spine extending beyond premaxillaries; eye moderate, 5.0 in head; inter- orbital 5.0; mouth large, nearly horizontal; lower jaw shorter than 5 The enumerations are only approximately correct because the fin rays are en- veloped in rather thick skin and are not all visible, and the scales are too small and the series too irregular to count accurately. NO. IO FISHES FROM PERU-—-HILDEBRAND AND BARTON Fic. 8—Lepophidium negropinna, new species. From the type, 280 mm. long, Talara, Pert (U.S.N.M. No. 144256). Loe = a VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS 30 ‘gjed ~winouops9g “yoelq Uul10} ‘lun [eue ‘oseq suoje soyojo[q yep pue us -Ieul Yep YIM [esi0g ‘ajed SI9AOD [[IS JO opIsuy ‘43302, prepurys ur Z°O1 ‘[eue JO UISIIO 0} AeCM}[eY ULY} d10ur A[qe -jOU SUIyOveI [210}09q *[B101 -oad jo aseq purtyeq 93]3 -}I] © [essop fo ulsiig *“solie][ixemoid jO ulZieur puosaq ivy Suipue}x9 = ‘jueurwoid A194 sutds = [e13s0x7 ‘4]3u2a] piepurjs ur $°Z yjydep sz ‘1ap “uss Aja}e1opom Apog *pedojaaap S ‘PrI+€ siayeI [IID 29 [esiop jo ulsii0 pue adeu usamjoq “Szz S@I19S [P19}e] UI seTeoS » VII ske1 [euy x BZI SABI [esiog sor1oeds Mau “nuurdosbau ‘uMoIq yiep wnouo0j19g ‘SUISILU Yep yoy ‘Aysnp [eue pue [esiog ‘uUMOIG yep SI9A0D [IS jo opisuy *4jSu2] prepuejs ur $°Z ‘dnd jo aajaureIp Aq Jeue JO UlSIIO YoRaI 0} SUI[Ie} ‘BuO, [e10}09q *[e1ojed jo aseq I9AO0 [eSIOp JO UISIID “S311 -[Ixemeid jo ulsieu qover 0} A[qejou Surrey ‘yzoys osuids [e1}Ss0yy *Y3u9s] Prepueys ur 6°S yidep szt ‘deep Apog *padojaaap 6 ‘91+ sioyeI [IID , 62 JesiIop jo ursrI0 pue sdeu usamjaq ‘S61 S9I19s [B19}e] UI SaTeOS » 08 Ske [euy yx ZOI SABI [eSIOg spjaunua ‘gjed wnou0z1eg ‘UISIeUL yIep snonulju0o & 4yIM [eue ‘SuluUIseq Iveu Yo}0[q yep e& YYWA [esiog ‘uMOIg yep SI9A0D [[IS JO ospisuy *4}3u9] piepuejs UI O°II ‘Jeue JO UTsIIO0 0} Aemyeq surygovor A[a01e0s [e10j0Ig *‘Jet0z}0ed JO Y}Sud[prur JIAO [eSIOp JO UISZIIO “solie[[Ixewoid JO UISIVUI puosdg [[eM sulyoeot ‘yuourm01d Joqze1 ouids [e1jsoy *y}su2e, pire -puejs ul 2°g yidap sir ‘jopus[s a}inb Apog *padojaaap Z *€1+% sioyer [ID »'9& JesIop jo uLsr10 pue odeu usamjoq ‘oz SO1IIS [P19}e] UI saTeoS , €0O1 shel [euy y 411 SAvi [esiod wmnisriqoubys ‘A[SULIIOUN JUNO 0} posueiie Al[ie[nZe111 pue [jews 00} 91¥ salvos 94} pUe “SIqISIA ][B JOU 91e PUB UIYS YOY} Joyjet ur podojaauo o1e sAeI UY 9} BSnedIq o}eINDDe jOU a1e SUOT}eIOUINUD sSOyT , ‘ajed wneou0p}18g ‘uIsieuL Yep snonuiju0d YM jeue ‘ulgieul uO saqd}0/q yep =6quM — s[esi0g ‘ated SI9A09 [[IS JO opisuy "q}3u2] pie -puejys ul €°g ‘jeue jo UIZIIO 0} ABMJ[eY ueyy sso] sulyovar [e10jIg *‘Je1oy0ed Jo y33u9| -piu =jO aoueApe ul ]]2" [esiop jo UIsIIGg *SOTIET -[Ixewoid jo ulsieu Sulyoeas ‘yuoutmo1d Ajarey aurds = [e.13s0y7 *yysus] pie -purys ul ’'g ydep sj ‘Japusjs ajinb Apog *pedojaaop b ‘9+ € siayer [IID »'8z [estlop jo utlst10 pue adeu useMmjeq ‘gzz SO119S [e19}e] UI SaTeoG ,/00I1 shel [euy » 411 SABI [esIOg ajppang ‘ajed wneu0j}i19g *‘Jesiop uo d]qISIA Ajpoieos ‘ulsieur y1ep YUM Ajizo11ajsod jeuy ‘ajed SI9A09 [JIS JO apisuy *y}su2] plepueys ur £°6 ‘jeue jo ulsiio 03 Aem -}[2Y Uey} s10m 194}e1 Sulyoeat [e10}09g *‘Jer0j00d jo Yq}sualprur 4J2A0 [eSIOp JO UIBIIO “solielpixeusid jO ulsiveu 0} Ajie3u sulqovat ‘yuouru01d you «© aurds [®44S0 yy *y}3u2] pie -puejys ur 1°24 yydep sz ‘dsap Alyelapour Apog *pedojaaap y 64% SsiayeI [IID , Sb jesiop jo ursI10 pue odeu usamjaq ‘ghz Sd119S [b19}e] UI SaTeIS , £01 shel [euy OZ SABI [es10(] $1gajosru ‘ajed «=wnouo0j}4I8g ‘SUISIVUL Yep yyiMm [eue pue jesiog ‘ajed SI9AOD []IS JO spIsuy 4332] piepur}s ul Sor ‘jeue JO UISZIIO 0} ABM -j[eY ueq}, sso] 19y4}e1 Suiygoeet [ei0jag *‘Tetojood jo 4}Sua] -plul JO a0ueApe UI 973 -}I] @ [estop Jo UlsIIO ‘SOLIE][Ixeusid jo us -1el puofkeq surqoeer Ajeo1eos ‘usurut01d j0u auids [e17S0y *q}8u2| piepue}s ut g°4 y3dep sz ‘tap -us[s A[a}e1apour Apog *padojaaap b 6+4€ siIOyeI [ID , O& [esiop jo ursIIO pue adeu usaMjoqg ‘z1z S9I1aS [B1d}e[ UI SaTeOS »'9OI Ske [euy » 421 SABI [esIog saqvsoagd NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON 31 the upper, included ; maxillary reaching well beyond posterior margin of eye, 2.1 in head; teeth in jaws in villiform bands, the outer ones in both jaws enlarged, those on vomer and palatines stronger, short and blunt; opercle with a concealed point; gill rakers 14 on lower limb, 5 somewhat developed, and 3 tubercles on upper limb; lateral line running rather high anteriorly, disappearing posteriorly; scales very small, rather difficult to enumerate accurately, extending forward on head to snout, 36 rows crossing back between nape and origin of dorsal; dorsal fin very long and low, the rays rather difficult to enu- merate because of heavy skin, the origin of fin a very short distance behind base of pectoral, its distance from tip of snout 4.0 in standard length; caudal fin scarcely as long as eye, slightly rounded, fully continuous with dorsal and anal; anal fin similar to dorsal, its origin behind that of dorsal a distance about equal to postorbital length of head ; the outer (or posterior) filament of ventral fin much the longer, 2.9 in head ; pectoral rather long, reaching notably more than halfway to origin of anal, 10.7 in standard length and 2.3 in head. Color uniform dark brown, scarcely paler below than above ; inside of gill covers pale; outer edge of dorsal fin black, base lighter with dark spots; caudal and anal fins blackish; ventral filaments pale ; pectoral a little darker than the general color of the body, with numerous dark punctulations visible under magnification; these punctulations present on the lower parts of the head and body. This apparently new Lepophidium is represented by a single speci- men (U.S.N.M. No. 144256), 280 mm. in total and 269 mm. in standard length. To identify this fish it was compared with the holotypes of all the species of the genus reported from off the Pacific coast of America. Its relationship with the other species, as indicated by the holotypes, is shown in the parallel comparison offered herewith. The numerous dorsal rays, the very small scales, the rather high number of gill rakers, the very prominent rostral spine, the anterior origin of the dorsal, the uniform dark brown body, and the black vertical fins characterize this species. The name, negropinna, was suggested by the black vertical fins. Genus OTOPHIDIUM Gill, 1885 Body elongate, considerably compressed ; head compressed ; snout without a spine; opercle with a strong spine; scales not in regular series and not imbricated (except in indefatigable, in which they are in fairly regular series and more or less imbricated), but partly or mostly at right angles to each other; dorsal and anal fins fully con- tinuous with the caudal. 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III This genus is said to differ from Ophidion in having a short thick air bladder with a large foramen. However, several species have not been examined for this character. OTOPHIDIUM FULVUM, new species FIGURE 9 Head 5.6; depth 8.6; D. about 115; A. about 83; P. 26 or 27. Body elongate, definitely compressed, its greatest thickness about two-thirds its depth; head rather deep, compressed ; snout somewhat pointed, no rostral spine, extending beyond premaxillaries, 5.7 in head ; eye slightly elongate, 3.4; interorbital very narrow, 12; mouth large, nearly horizontal ; lower jaw shorter than the upper, included ; maxillary extending well beyond posterior margin of pupil, 2.4 in head; teeth in jaws in villiform bands, the outer ones in each jaw somewhat enlarged, those on vomer and palatines strong, low, and bluntly pointed; opercle with a strong spine; gill rakers consisting of spiny tubercles, 4 on the lower limb (2 somewhat developed), and 2 on the upper one; lateral line rather high, absent posteriorly ; scales not in regular series, more or less embedded, elongate, many at right angles to each other; dorsal very long and low, the rays difficult to enumerate, origin of fin well behind midlength of pectoral, its distance from tip of snout 4.2 in standard length; caudal fin very short, con- tinuous with the dorsal and anal; anal fin similar to the dorsal, though shorter, its origin about length of head behind that of dorsal, and its distance from tip of mandible 2.8 in standard length; ventral inserted at vertical from middle of eye, the outer filament the longer, 2.0 in head ; pectoral fin short and broad, with rounded margin, with many rays,® reaching notably less than halfway to origin of anal, 11.6 in standard length, 2.05 in head. Color uniform brown, though somewhat lighter on chest and ab- domen than elsewhere; many dark punctulations visible on the body under magnification ; dorsal and anal fins with narrow dark margins, extending around the tail. This apparently new species is represented by a single specimen (U.S.N.M. No. 144257) 72 mm. in total and 69 mm. in standard length. It seems to be nearest O. galeoides Gilbert, known from the Gulf of California, the type of which is at hand. From that species’ it differs, however, in several characters, as shown in the parallel comparison offered herewith. From indefatigable Jordan and Boll- $To obtain an accurate enumeration of the rays the skin was cut behind the fin and near the base. NO. IO FISHES FROM PERU—-HILDEBRAND AND BARTON Fic. 9.—Otophidium fulvum, new species. From the type, 72 mm. long, Talara, Pertti (U.S.N.M. No. 144257). 33 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III man, which is known from Panama Bay to the Galapagos Islands, it differs prominently in the arrangement of the scales, which are in irregular series and at right angles to each other in fulvum, while they are nearly all in regular series and more or less imbricated in indefatigable. It differs prominently from O. scrippsi Hubbs in the fewer, shorter, spiny gill rakers, and slenderer body. The specific name fulvum is in allusion to the brown color of the specimen described. galeoides Pectoral fin moderate, reaching nota- bly more than halfway to origin of anal, 7.25 in standard length, and 1.35 in head, with 21 rays. Gill rakers strong, 4 on lower limb all fairly well developed, and 2 tuber- cles on upper limb of first arch. Maxillary reaching posterior margin of pupil, 2.15 in head. Origin of dorsal over midlength of fulvum, new species Pectoral fin shorter, reaching notably less than halfway to origin of anal, 11.6 in standard length, and 2.05 in head, with 26 or 27 rays. Gill rakers all spiny tubercles, none well developed, 4 on lower limb and 2 on the upper one of first arch. Maxillary reaching a little beyond posterior margin of pupil, 2.4 in head. Origin of dorsal well behind mid- length of pectoral, its distance from _ tip of snout 4.2 in standard length. pectoral, its distance from tip of snout 3.75 in standard length. Family MUGILIDAE: Mullets MUGIL CEPHALUS Linnaeus Mugil cephalus H1LpEBRAND, 1946, p. 422 (description). Two small specimens, 30 and 43 mm. in total and 24 and 33 mm. in standard length, are included in the collections. The juvenile anal fin formula, II, 9, has been retained in these specimens. M. cephalus seems to be a common species in northern Pert. Numerous specimens of young and a few adults were included in the collections made by the U. S. Fish and Wildlife Service mission to Pert in 1941 (Hilde- brand, 1946, p. 422). Range.—Shores of nearly all warm seas; on the Pacific coast of America from California to Chile. MUGIL CUREMA Cuvier and Valenciennes Mugil curema HitbeBranp, 10946, p. 426 (description). One small specimen 45 mm. in total and 37 mm. in standard length is included in the collection. This specimen still has the juvenile anal NO. I0 FISHES FROM PERU-—-HILDEBRAND AND BARTON 35 fin formula, II, 10, the third ray having not yet developed into a spine. Only a few young and no adults were included in the collec- tions made by the U. S. Fish and Wildlife Service mission to Pert in 1941 (Hildebrand, 1946, p. 426). Range.—Known from both coasts of America; on the Pacific coast from the Gulf of California to Chile. LITERATURE CITED BarRTON, OTIs. 1947. Two new species, an Eques and a Holanthias, from Peru. Amer. Mus. Novitates, No. 1350, 3 pp., 2 figs. Crark, Howarp WALTON. 1936. The Templeton Crocker expedition of the California Academy of Sciences, 1932, new and noteworthy fishes. Proc. California Acad. Sci., ser. 4, vol. 21, No. 20, pp. 383-306. Cuvier, GEORGES, and VALENCIENNES, ACHILLE. 1828-1840. Histoire naturelle des poissons, 22 vols., 650 pls. GARMAN, SAMUEL. 1899. Reports on an exploration off the west coasts of Mexico, Central and South America, and off the Galapagos Islands, in charge of Alexander Agassiz by the U. S. Fish Commission Steamer Alba- tross during 1891 (Lt. Comdr. Z. L. Tanner, U. S. N., command- ing). Mem. Mus. Comp. Zool., vol. 24, 431 pp., 97 pls. GILBERT, CHARLES HENRY. 1890. Scientific results of explorations by the U. S. Fish Commission Steamer Albatross. No. 12. A preliminary report on the fishes collected by the Steamer Albatross on the Pacific coast of North America during the year 1889, with descriptions of twelve new genera and ninety-two new species. Proc. U. S. Nat. Mus., vol. 13, PP. 49-126. GILL, THEODORE NICHOLAS. 1862. Catalogue of the fishes of Lower California in the Smithsonian Institution, collected by Mr. J. Xantus. Proc. Acad. Nat. Sci. Philadelphia, 1862, pp. 140-151 (pt. 1); pp. 242-246 (pt. 2); pp. 249-262 (pt. 3). HILpEBRAND, SAMUEL FREDERICK. 1946. A descriptive catalog of the shore fishes of Peru. U. S. Nat. Mus. Bull. 180, 530 pp., 95 figs. Husss, Cart LEAVITT. 1944. Species of the circumtropical fish genus Brotula. Copeia, 1944, No. 3, pp. 162-178. 36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III JorpAN, Davip STARR, and EvERMANN, BARTON WARREN. 1896-1900. The fishes of North and Middle America. U. S. Nat. Mus. Bull. 47, pt. 1, 1896, pp. i-lx, 1-1240; pt. 2, 1898, pp. i-xxx, 1241-2183; pt. 3, 1808, pp. i-xxiv, 2183a-3136; pt. 4, 1900, pp. i-ci, 3137-3313, 292 pls. LINNAEUS, CARL. 1758. Systema naturae, ed. 10, 824 pp. 1766. Systema naturae, ed. 12, vol. I, 532 pp. MEEK, SETH EuGENE, and HILpEBRAND, SAMUEL FREDERICK. 1923-1928. The marine fishes of Panama. Publ. Field Mus. Nat. Hist., zool. ser., vol. 15, pt. I, 1923, pp. I-330, pls. 1-24; pt. 2, 1925, pp. 331-707, pls. 25-71; pt. 3, 1928, pp. 709-1045, pls. 72-102. Myers, GEORGE SPRAGUE, and WaAbDE, CHARLES BARKLEY. 1941. Four new genera and ten new species of eels from the Pacific coast of tropical America. Allan Hancock Pacific Expeditions, vol. 9, No. 4, pp. 65-111, pls. 7-16. NorMAN, JOHN ROXBROUGH. 1937. Coast fishes. The Patagonian region. Discovery Reports, vol. 16, pt. 2, 150 pp., 76 text figs., 5 pls. RAFINESQUE, CONSTANTINE SAMUEL. 1815. Analyse de la nature, au tableau de l’univers et des corps organises, 224 pp. Palerme. SEALE, ALVIN. 1940. Report on fishes from Allen Hancock Expeditions in the California Academy of Sciences. Allan Hancock Pacific Expeditions, vol. 9, No. I, pp. 1-46, pls. 1-5. VALENCIENNES, ACHILLE. 1855. Voyage autour du monde sur la frégate Venus: Zoologie, 351 pp. (Ichthyologie, pp. 297-351). (10 pls. published in 1846, with some differences in nomenclature. ), Wabde, CHARLES BARKLEY. 1946. New fishes in the collections of the Allan Hancock Foundation. Allan Hancock Pacific Expeditions, vol. 9, No. 8, pp. 215-237, pls. 29-32. ee \Y Guay T. i mls nae =e 1 ia -SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 11 Chomas Lincoln Casey Fund. “ZARVAE OF ‘THE ELATERID BEETLES OR THE TRIBE’ LEPTUROIOINI (COLEOPTERA: ELATERIDAE) BY ROBERT GLEN 1 ‘ Site's hg ae r Research Coordinator, Division of Entomology, Science Service, Department of Agriculture, | Ottawa, Canada (PUBLICATION 3987) CITY OF WASHINGTON eer BY THE SMITHSONIAN INSTITUTION APRIL 19, an SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 11 Thomas Lincoln Casey Fund Peon Ae OF The ELATERID BEETLES OFF m=) TRIBE LEPTUROIDINI (GOEL OP TERA EEATERIDAE) BY ROBERT GLEN Research Coordinator, Division of Entomology, Science Service, Department of Agriculture, Ottawa, Canada Ss at INCRE y Vo LOS Seen Ww. A a < (PUBLICATION 3987 ) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION APRIL 19, 1950 Ae The Lord Baltimore Press BALTIMORE, MD., U. & CONTENTS Tsingtao (OCS HOY DT ec eel coh Rene Nc Rca RRR AE try aE ee SRN area PRS Son Sate /NGITAUON ANT Fever STU Ph tes ceca ROR ROR EEG SIE cioe acinar nnes Acomeosbac IB Gerben cmc oer ieee tere or farasronsiel dis) w ial ounstter ok jasaaerevae Acfalaolotereaaielt t,o arene Niethodsgancdeprocedunenmecate ns dc crisiartacareee obi acme accent RCE AGGIE] UMP Me Tan Teer sens Cae’ sche eT eiaie Ore GPSS ST eso aye vesex aes icra ere aa Bresenvatlonandestondee Of itatentialls jaye qciacteiiciisc cr cieieia ateteicrie cies Preparatongoteiateniall of examination asi cclacieicreciie amici rice RE amMinat OMmOtmInateisial eae serreoceniicrcee nec e eno te sihemreconnalssancemsitvieynn oe ccieseeciinianis ie aici ition a creer selremcl eta] CCS EtG Vises ceoitaiev top wovetatore tetotods .c tasl oes Sue Talele eas ceed ey Se NODE VAAHONS. Used iM TllastratiOns. «0. cc sac ocis-- Euadiws hierogly phicusm (Say) ee secs eee seein eee LEUCis) (PUdiCuUsy USrO Wiessner Ne ese aioke ecto evel lefode eet ANN JOOS ae aMlOn NS FATOEIT oo ooo bedosoShonoonodn00ORnOoS JLgnhins igaimclkilenns (GQREGGCEND) .o5cnohosng0pdeccceousccaec JETS (AoE 2CSES (Ue) (B)ecceccosngecsoooeooucns iv CONTENTS Page The Ludius fallax SroOup. canes c wer acme ele leet rere 82 Ludius tessellatus (Linnaeus) «..%2.: wilco «ts seeeteretein eee 84 Ludius castaneus (uinnaeus) as.ce oe oto 89 Ludiusviduus Brown" 2) asecese etree eee eee go Ludius bombycinus’ (German)... 6.) 2.202 nlie eee OI Ludvus medianus. (Germat )..!. 0... vs. 5020s ae eee eee 92 Ludwus purpureus (Poda) .... . 3.0%) oc see eee 93 Ludius affinis (Paykull) «2.62 6/0 sac 0100 esi 5 si 93 The Ludwusirotundicollis Sroups eee aici enter nee O04 Ludius rotundicollis (Say... +. «+2 0=.4.s=yee eee 96 Ludwus sulcicollts (Say) 0 v2. < oc einen see oe eee 102 Ludiusemctus (Paykull) vu.22 vee as eee 104 Ludius bipusiulaius (Linnaeus) ) ; mandibles with retinac- ulum; nasale well developed................. Subfamily ELATERINAE 4a, IHoRiimsaianan jaeioenleye (Piz 1s, Oy ©))ooo0eb6000 00500 0ne00G0a0000p4dr 5 Postmentum subrectangular (pit, fig. 9, h)..Tribe LEPTUROIDINI (p. 33) 5. Mandibles without teeth on inner aspect (fig. 9, d, e); tenth ab- dominal segment usually with “anal armature” (ar, fig. 9, Ce ocicse 6 GIS READ Cory OIE TECH ERRES EET e Re SAR o Tribe PYROPHORINI Mandibles with 3 teeth on inner aspect...............-..- Tribe PITYOBINI LARVAL CHARACTERISTICS OF THE TRIBE LEPTUROIDINI On the basis of larval characters, the tribe Lepturoidini includes the species separated by Leng (1920) into the tribes Lepturoidini, Hypnoidini, and Melanactini. Hyslop (1917) characterized the tribe as follows: Postmentum broad caudad, mandibles with teeth on inner surface, and tenth abdominal segment without armature. There is no doubt that these are the primary diagnostic characters of this group. However, a more nearly complete characterization of a lepturoidine larva would be: Ninth abdominal segment emarginate posteriorly (cn, figs. 8, d; 10, a, f; 11, a) and without dorsal prongs anterad to urogomphi; tenth abdominal segment without armature; the bases of stipites distinctly separated and postmentum subrectangular (pmt, fig. 9, h) ; mandibles with I or more teeth on inner surface (figs. 2, e; 9, f, g); nasale well developed (m, figs. 9, 7; 19, b); gula present (gu, figs. 1, b; 19, f; 27, b), sometimes very narrow; and cardines moderately large (cd, figs. 0; h; 22, b).. 3In larvae of the genus Hemirhipus (q.v. Boving and Craighead, 1931, pl. 84 G) the urogomphi are fused except at the tips, resulting in a small but distinct notch. In some first-instar larvae this notch may be closed (op. cit. pl. 85 O), but the line of fusion of the urogomphi is evident. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOR. Eat , KEY TO GENERA, ISOLATED “SPECIES GROUPS,” AND ISOLATED SPECIES OF THE TRIBE LEPTUROIDINI 1. Urogomphi undivided’ (aw; fig. 10} /a, D0) 20.5. yokes ee erent 2 Urogomphi divided: prongs subequal (figs. 10, f; 11, f; 12, a, f), or unequal (figs: DT, G5 12; (By %6,0€) a. es wom .eicrsohte keke See eee 3 2. Urogomphi blunt (fig. 10, a); abdominal pleurites absent (fig. 10, CDSE PERERA ols lcs aoe a eee Eanus (p. 187) Urogomphi sharp (fig. 10, b) ; abdominal pleurites large. Hypnoidus (p. 186) 3. Head bearing dorsal posteroepicranial setae (ped, fig. 10, g); thoracic segments and first 8 abdominal segments bearing medial antero- tergal ‘setae: i(@im, fig: 10, g)\s isn. a. bo Son Se ee ee 4 Without dorsal posteroepicranial setae; usually without medial antero- tergal setae auc Gene aie dsiaie bie ctdvele a love Soke ePNRDS LLU 2 a ee 7 4. Ninth abdominal segment (fig. 10, f) with sharp “teeth” (to) on sides of dorsum and 2 setae on central dorsal area; Aus- iialiat sects ere eae Crepidomenus queenslandicus Blair (p. 181) Ninth abdominal segment with well-rounded “teeth” on sides of dorsum ands4usetaczonucentraladorsal areas. eee rere ener nner ena 5 5. Abdominal mediotergites (as in fig. 12, d) with conspicuous trans- verse rugae (crescent-shaped shallow pits, which are sometimes confluent) ; eyes absent; larvae attaining more than 30 mm. in lenethiay tg Pek Meee cee nye Melanactes densus LeConte (p. 189) Abdominal mediotergites without such sculpturing; eyes present....... 6 6. Abdominal mediotergites with transverse branch of impressions extend- ing to or practically to the mediodorsal suture on second to eighth SERITIEMES tes menor Is cut nee cheer oe Ludius, the nitidulus group (p. 111) Abdominal mediotergites with transverse branch of impressions extend- ing approximately one-half the distance from the longitudinal branch to the mediodorsal suture............. Cryptohypnus (p. 183) 7. Spiracles in eighth abdominal segment (sf, fig. 11, e) twice as long as spiracles in seventh abdominal segment................+..ssee0<: poe Bepries ris) ae mes eee RENT NE cries Ludius, the pyrrhos group (p. 147) Spiracles subequal in size in seventh and eighth abdominal segments.... 8 8. Presternum of prothorax divided into 2 or more sclerites (prst, fig. NORD COM I tas Sota DOR ME IOS Soa On Ludius, pars majora (p. 35) Presternum of prothorax undivided, of 1 large triangular sclerite (prst, AS WON) 5 ee ova d othe nucle tie be edad se eee eee 9 9. (Caudal notchismall (figs. 11, 1a, fi; 12,0). tance ee eee 10 Gatidal-notch large ((fig! 12; ef )inl. see ee oc ee 16 10. Outer prongs of urogomphi reduced to mere tubercles (opr, fig. 11, (ISH fh) aa NSE A eRe RYN: Sele emcee Se pote Per 4) Seat Limonius, pars (p. 157) Outer prongs of urogomphi definitely pronglike (opr, figs. 11, b, g; 7) ee I ea cee en et ERATE Na II 11. Outer prongs of urogomphi not longer than inner prongs (fig. 11, f-h). 12 Outer prongs of urogomphi much longer than inner prongs (fig. 12, DB) BS a Paha eas old sghele ele lbaa eae ale alee Ne eee 15 12. Ninth abdominal segment with a distinct (sometimes short) mediodorsal groove (mg, fig. 11, f); eyes present; North America.......... 13 Ninth abdominal segment without a mediodorsal groove (fig. 11, g, h); eyes absent in North American species. ....252 .4-s eee 14 NO. II LARVAE OF THE ELATERID BEETLES—GLEN 35 13. Outer prongs of urogomphi (opr, fig. 11, b, f) projected caudodorsad and usually slightly laterad, not curving anteriorly, tip blunt; ab- dominal mediotergites usually with impressions extending to the mediodorsal suture in second to fifth segments................... Sid et eb ts ener SPE nl ea Ludius resplendens aerarius (Randall) (p. 136) Outer prongs of urogomphi (opr, fig. 11, c) projected dorsad, curving anteriorly, tip sharp; abdominal mediotergites with impressions definitely not reaching to mediodorsal suture.................-00- RORY oh ees Uneven chiscanene dvonaiaieeee oaks Elathous bicolor (LeConte) (p. 167) 14. Ninth abdominal segment (fig. 11, g) with well-rounded “teeth” (to) Om Giles Ge GOR le oo ocnoabdcoocusavonancce Limonius, pars (p. 157) Ninth abdominal segment (fig. 11, 1) with prominent pointed “teeth” (to) on sides of dorsum....Ludius, the limoniiformis group (p. 150) 15. Dorsum very dark, usually dark reddish brown to brownish black; abdominal mediotergites punctulate, but without transverse rugae; inner prongs of urogomphi (ipr, fig. 12, b) smooth, without pos- US GUOF CHE OS) ge) (hake ne eed cee eRe ee eae Ae Lepturoides (p. 168) Dorsum never dark, usually yellow to yellowish brown. Athous, pars (p. 170) 16. Ninth abdominal segment (fig. 12, e) without a mediodorsal groove, with four setae on central dorsal area, and with very long outer urogomphal prongs (opr) ; without impressions on mediotergites of mesothorax and metathorax..Ludius divaricatus (LeConte) (p. 118) Ninth abdominal segment (fig. 12, f) with a mediodorsal groove (mg), and without setae on central dorsal area; with definite impressions on mediotergites of mesothorax and metathorax.................. 17 17. Eyes absent; abdominal mediotergites (fig. 12, d) with prominent trans- verse rugae (crescent-shaped, shallow pits which are sometimes confluent) ; urogomphal prongs (fig. 12, f) subequal, outer prongs pombeda whetia tmeroded iis Aimcrcate oaths we be = As toate ove sree cree ae Soo TCO: Hemicrepidius, and Athous niger and its allies (p. 178) Eyes present; American species with prominent pits or rugae, Euro- pean species sparsely punctulate; outer urogomphal prongs either with bluntly rounded tips (opr, fig. 12, a) or much longer than inner DRONSTS WO TC)! oy sice he eoieae me oee eee ose as Athous, pars (p. 170) Genus LUDIUS Eschscholtz 4 IPIGURES 1-7/0) GO; h, 7s LO; ds U1, b,c; f, he We, @5 13-20 For many years nomenclatorial confusion has involved the insects now recognized under the generic name of Ludius Eschscholtz. Re- ferring to this situation, Hyslop (1921, p. 621) states, “.. . the genus Elater, as recognized by contemporary coleopterists, is in reality the genus Ampedus, the insects now recognized under the 4Recent publications by Dietrich (1945, p. 19) and Lane (1948, p. 182) suggest that Ludius Eschscholtz should be placed in synonomy under Ctenicera Latreille. (Vide: Henry Dietrich, “The Elateridae of New York State,” Mem. 269, Cornell Univ. Agr. Exper. Stat., Ithaca, N. Y., Jan. 1945; M. C. Lane, “Some Generic Corrections in the Elateridae I,” Proc. Ent. Soc. Washington, vol. 50, No. 7, pp. 179-182, Oct. 1948.) 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III generic name Ludius being truly Elater. Ludius, on the other hand, should be applied to the insects generally known as Corymbites, and Corymbites as a generic name disappears.” North American writers, in general, follow Hyslop and use Ludius instead of Corymbites, but European coleopterists still cling to the latter or to one or another of the subgenera recognized by Schenkling (1927). In the present study the name Ludius Eschscholtz includes all species of the fol- lowing generic and subgeneric names: Actenicerus Kiesenwetter, Anostirus Thomson, Aphotistus Kirby, Calambus Thomson, Corym- bites Latreille, Ctenicera Latreille, Ctenicerus Stephens, Diacanthus Latreille, Haplotarsus Stephens, Hypoganus Kiesenwetter, Liotrichus Kiesenwetter, Prosternon Latreille, Selatosomus Stephens, and Tac- tocomus Kiesenwetter. Where possible, the names proposed by Brown (1935, 1936) for various “species groups” of Ludius have been used. Knowledge of the larvae of the genus Ludius is based upon 44 species, 29 occurring in North America and 16 in Eurasia, one being common to both regions. All these, except 6 Eurasian species, have been examined in the present study. The genus is extremely diverse, both in habit and in structure. Larvae occur normally in soil, decaying wood, and forest litter, but specimens have been found also in cow dung, mushrooms, and sphag- num, and under stones. Some species prefer dry situations, others re- quire abundant moisture. In the world as a whole, Ludius is the most destructive genus in the Elateridae. But even the most phytophagous species seem capable of sustaining themselves for periods of months or years on decomposing organic matter, or of reverting to predaceous habits when opportunity or necessity arises. Some species are chiefly entomophagous and have never been associated with damage to culti- vated plants. Larvae have been known to attack spiders, the active and inactive stages of various insects, and even members of their own species. With the exception of a few pest species, the life history is either unknown or very incompletely known. On the basis of the informa- tion available, the common normal life cycle is as follows: The adults mate and lay their eggs in May and June; the eggs hatch within a few weeks; the larvae live for 3 or more years, pupating, when mature, in July or August; the adults develop within a month after pupation and remain in their pupal chambers until the following spring. Pupation occurs in June with some species, and undoubtedly other exceptions will be found as our knowledge of the biology of NOP el LARVAE OF THE ELATERID BEETLES GLEN 37 the group increases. Only one record has been found in the literature to date of any species of Ludius (L. amplicollis) normally completing its life cycle within 1 year. Concerning the structural features of the genus Ludius, Henriksen (1911, p. 258) states (translation) : “A genus which it is very diffi- cult to define, as the known larvae show such variation in charac- ters.” The writer agrees fully with this statement. So much hetero- geneity exists in the larvae assembled that no single character or combination of characters has been found which adequately defines the assemblage as a natural group. However, all Ludius larvae examined have the urogomphi bifid and are thus distinguished from Hypnoidus and Eanus. Separation from Cryptohypnus and Crepidomenus is obtained for all known Ludius, except the nitidulus group, through the absence of dorsal posteroepicranial setae (ped, fig. 10, g). With the exception of seven species (divaricatus, resplendens, sjaelandicus, pyrrhos, protractus, limoniiformis, and cylindriformis), all the Ludius examined have the prosternum divided and are thus separable from the larvae of Limonius, Elathous, Lepturoides, Athous, Hemicrepidius, and Mela- nactes. Of these seven species, divaricatus is quite isolated from other known Lepturoidini, and the other six probably are annectant species, as mentioned by Van Dyke (1932, p. 389), connecting Ludius with Limonius and with the Athous complex. Because of these findings the writer feels justified in stressing the taxonomic value of characters of the prosternum. By combining the characters of the caudal notch and the prosternum, four major groups of Ludius have been established. These form a good working basis for the identification of the larvae of this genus. Group I: Caudal notch small (fig. 14, b, g); prosternum divided into two or more sclerites (figs. 10, d; 13, a). This combination of characters is unique in the known Lepturoidini with the exception of Eanus, which is readily distinguished by the undivided urogomphi. This group includes the cupreus group, appressus (Randall), and sjaelandicus (Muller). Group II: Caudal notch large (figs. 13, 1; 14, c, 1, 7) ; prosternum divided into two or more sclerites (figs. 10, d; 13, a). Cryptohypnus and Crepidomenus larvae also possess these characters, but are dis- tinguished by a combination of sculptural and setal patterns. This group includes two-thirds of the species of Ludius known in the larval stage: aeripennis group, inflatus group, edwardsi group, semi- vittatus (Say), propola group, triundulatus group, fallax group, rotundicollis group, bipustulatus (Linnaeus), and nitidulus group. 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLe2 lint Group III: Caudal notch small (fig. 11, f, h); prosternum un- divided, consisting of one triangular sclerite (fig. 13, b) sometimes deeply incised on posterolateral aspects. This combination of char- acters is found in all species of Limonius, Elathous, and Lepturoides, and in many species of Athous. Proper characterization of the Ludius that fall in this group can be made only on the basis of “species group” characters as given in the key to genera, isolated species groups, and isolated species of the tribe Lepturoidini (p. 34) or as discussed under the species groups concerned: pyrrhos group, limoniuformis group, and resplendens aerarius (Randall). | Group IV: Caudal notch large (fig. 12, e) ; prosternum undivided, consisting of one large triangular sclerite, much as represented in figures 10, e and 13, b. These characters are found in Hypnoidus, Hemicrepidius, Melanactes, and in some Athous. The only Ludius falling in this group is divaricatus (LeConte) and it is readily sepa- rated from other known Lepturoidini by the characteristic ninth abdominal segment (fig. 24, c, d). Considering the genus in the broad sense in which it is currently recognized, the mature larvae may be described as follows, characters common to all species being marked with an asterisk: Length 14 to 30 mm. Shape varying from the robust pectinicornis, measuring 28 mm. by 4 mm., to the relatively slender cylindriformis, measuring 30 mm. by 2.6 mm. Body usually widest across fourth abdominal segment, but in some species widest across thorax and first 2 abdominal segments; some species very robust with relatively small pleural membranes, others flattened with conspicuous large pleural areas. Dorsum pale yellow to very dark brown; without color pattern or distinctly patterned. *Urogomphi always bifid. Uro- gomphal prongs varying tremendously as to shape and relative length. Caudal notch large or small; posterior aperture varying from wide to almost closed. *Dorsum of ninth abdominal segment always with “teeth” on lateral margins. Dorsal plate of ninth abdominal segment varying from very convex to almost flat, the following characters being present or absent: setae on central area, large pits, median groove, and transverse impression. Frontoclypeal region reaching to foramen magnum or failing to attain foramen; terminating trun- cate, broadly rounded, or bluntly pointed. Nasale of one tooth termi- nating in one or three points. Subnasale highly variable. Eyes present (usually) or absent; from 2 to 6 setae around each eye or eye region. From 2 to 7 lateroepicranial setae in unpaired or paired arrangement. Each dorsal head sulcus with 4 or 5 setae, sometimes additional minute setae. Gula short and wide, or short and narrow, or elongate and NO. II LARVAE OF THE ELATERID BEETLES—GLEN 39 narrow. *Antennae with basal segment longest and terminal seg- ment shortest. One to 6 “sensory” appendices on second segment of antenna. *Mandibles with well-developed retinaculum and without other prominent teeth. Proxistipes and dististipes not distinct except in rotundicollis group. From 2 to 8 prominent setae on antero-latero- ventral aspect of stipes. *Second segment of maxillary palpi without setae. Postmentum with 1 or 2 setae at each corner. *Second seg- ment of labial palpi not longer than first segment. Presternum of prothorax of 1, 2, 3, or 4 sclerites. Each episternum of mesothorax and metathorax without spinelike setae or bearing from I to ro such setae. Legs bearing many spinelike setae, number and arrangement variable. “Abdominal mediotergites (segments I to 8 inclusive) usu- ally punctulate without prominent transverse rugae or large deep pits. Abdominal mediotergites bearing “impressions,” length vari- able; setal pattern highly variable in both number and arrangement. Abdominal pleurites vary from “almost wanting” to well developed. Sternum in first to eighth abdominal segments either of I or 3 sclerites. Abdominal spiracles usually subequal and in anterior half of each segment; sometimes in posterior part of eighth segment; sometimes enlarged in eighth segment. Tenth abdominal segment with anal aperture linear or T-shaped. KEY TO “SPECIES GROUPS” AND ISOLATED SPECIES OF LUDIUS ® TeGatdal moteh large Ghos, 19; Wy 45° 04,6) atin. LUDIUS SJAELANDICUS (Miiller) FicurEsS 14, f; 27 Elater sjaelandicus MU.ier, Fauna Insectorum Friedrichsdalina, p. 21, 1764. Elater tessellatus Fasricrtus (nec Linnaeus), Systema Eleutheratorum, p. 211, 1775. Corymbites tessellatus (Fabricius), CANDEZE, Monographie des elatérides, vol. 4, p. 104, 1863. Corymbites sjaelandicus (Muller), Scutoptr, Naturh. Tidsskr., ser. 3, vol. 3, p. 554, 1865. Corymbites (Actenicerus) sjaelandicus (Miller), SCHENKLING, Coleopt. Cat. (ed. Junk), vol. 2, pt. 88, p. 368, 1927. Ludius sjaelandicus (Miller), VAN Dyke, Proc. California Acad. Sci., vol. 20, Pp. 391, 1932. This species is well known in northern and central Europe and in Siberia. According to Leng (1920, p. 169) and Van Dyke (1932, NO. If LARVAE OF THE ELATERID BEETLES—GLEN 143 p. 391) sjaelandicus also inhabits the northeastern part of North America, but no larvae have been obtained from this region. Larvae have been collected from rotten birch, sphagnum, and cow dung, but the typical habitat is the moist soil of low-lying land such as peat moors, alder swamps, grassy meadows, and damp, moss- covered parts of forests. Injury to domestic plants frequently results when such areas are brought under cultivation. Durnovo (1935) reports this species as a pest of vegetables on peaty soil recently cleared of alder. He further states that the larvae are much less numerous in clayey soils and practically absent from sandy areas. Pupation occurs in late July or early August. The larva has been described from Denmark by Schiodte (1870, p. 521) and Henriksen (1911, pp. 261-262), and from Germany by Beling (1883, pp. 272-273, “tessellatus L.”). It is readily recognized by having the spiracles situated in the posterior half of the eighth abdominal segment (sp, fig. 27, d), but not enlarged as in the pyrrhos group. The larva is further characterized by its bright yellow color ; tridentate nasale (n, fig. 27, a) ; presternum of prothorax divided into 3 parts; ninth abdominal segment (fig. 27, c, d) with small caudal notch (cn), urogomphi (ur) bifid with inner prongs (7pr) slightly longer and much more robust than outer prongs (opr), 2 setae on central dorsal area, and 3 prominent blunt teeth (to) on lateral mar- gins of dorsum. Description of “mature” larva.—Length 22 mm.; greatest breadth 3.0 mm. on fourth and fifth abdominal segments. Henriksen (1911, p. 262) and Beling (1883, p. 272) record specimens measuring 26 mm. in length. Body robust; with moderately large membranes on lateral aspect; all segments broader than long; head and ninth abdominal segment about three-fourths greatest body width. Dorsum bright yellow (near “hazel” or slightly darker than “ochraceous buff,’ Ridg- way, 1912) ; mandibles, nasale, talus, and prongs of urogomphi darker ; venter slightly paler. Dorsum shiny; slightly rugose; with a few small and very shallow inconspicuous pits. Head subquadrangular with arcuate sides, about as thick at base as long; flattened above and below. Frontoclypeal region with posterior part extending backward almost to foramen magnum, truncate posteriorly. Two prominent anterior nasosulcal setae on each side of base of nasale. Nasale (n, fig. 27, a) with tip tridentate when uneroded; median denticle largest, lateral denticles projecting anterolaterad. Subnasale (sm) consisting of strongly sclerotized transverse ridge with large, forward-projecting 144 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III median denticle; finely serrate (when uneroded) on each side of median tooth. Paranasal lobes produced beyond nasale, each bearing 3 setae (1 small). Epicranial plates nearly smooth. Dorsal sulci shallow, each with 4 setae subequally spaced, the most anterior seta being very long; others very small, somewhat peglike. Ventral sulci bearing row of 2 to 4 setae, only I or 2 conspicuous. Two large unpaired latero- epicranial setae. Eye spot black, well defined, ovate or circular; bordered by 2 prominent setae. Postgenal areas expanded mesad, almost or actually meeting. Gula (gu, fig. 27, b) short, greatly narrowed (sometimes oblit- erated) in the middle by converging postoccipital sutures; glabrous. Antenna with first joint clavate, one-half to three-fifths as wide as long ; without setae; 7 to g small pores. Second joint subcylindri- cal, slightly more than one-half as wide as long; two-thirds length of basal joint ; 1 or 2 pores; a few minute setae borne distally ; 1 medium- sized conical “sensory” appendix just ventrad to base of third joint. Terminal segment small, about one-third as long as second segment and one-quarter as wide; 4 setae on apex. Mandible of moderate length, robust; ventrally, about two-thirds as wide at base (ventral aspect) as long; retinaculum very large; penicillus sometimes reaching base of retinaculum. Distal half inward bending ; pointed ; outer surface convex with shallow dorsal groove ; inner face slightly excavate with small median carina, ventral margin of inner face sharp and slightly convex ventrally, dorsal margin sharp and more strongly convex dorsally. Ventral mouthparts three-fourths as wide across bases of stipites as across anterior ends of stipites. Cardines slightly separated ; without seta between rami of Y-shaped brace. Stipes large, subrectangular ; proxistipes and dististipes not distinct; with 2 prominent setae on antero-lateroventral aspect. Galea with basal joint subcylindrical, about as long as terminal joint, without setae or pores ; terminal seg- ment narrower than basal segment, with 4 to 6 pores on lateroventral aspect. Maxillary palpi with all segments subcylindrical. First joint almost as wide as long; distally on mesoventral surface with group of 3 to 5 small pores; without setae. Second segment almost twice as long as wide; longer than first segment and as wide; without setae ; with 3 or 4 pores. Third segment almost as long as wide; one-half or less length of second segment; without setae; with 2 or 3 pores ventrally. Fourth segment slightly longer than wide; almost as long as third segment; without pores or setae. Postmentum with 1 long NO. II LARVAE OF THE ELATERID BEETLES—GLEN 145 seta at each corner. First prementum with 1 large seta just caudad to base of each palpus. Labial palpus with basal joint at least one-half length of first prementum, longer than wide, without setae, with 5 or 6 pores; terminal joint about one-half length and less than one-half width of basal segment, without setae, usually with 1 pore. Prothorax nearly equal to combined length of mesothorax and metathorax; wider posteriorly; slightly wider than long. Tergites slightly rugose, with few small and very shallow pits; anteriorly with 2 pairs large setae (on each side of median dorsal suture) in trans- verse row; posteriorly with 4 prominent setae in transverse row, only 2 most medial setae paired; glabrous elsewhere. Episternum with 1 large seta. Epimeron bearing I small seta. Presternal area consist- ing of 3 sclerites as follows: A small posterior median sclerite, ante- riorly attenuate; 2 large subtriangular lateral sclerites (sometimes partially fused) striate on anterolateral aspect, with 1 stout seta later- ally and a diagonal row of 4 or 5 minute setae on anteromedial aspect. Eusternum small, usually with narrow medial sclerite. Sternellum and poststernellum indefinite, small, membranous. Mesothorax and metathorax each about twice as wide as long. Mediotergites with few small and very shallow pits ; transverse branch of impression reaching about one-fourth distance from longitudinal branch to middorsal suture ; longitudinal branch of impression short. Anterior part of each mediotergite with 1 or 2 minute setae laterad to impression. Posterior part of mediotergite with 4 conspicuous setae in transverse row, only the 2 most median setae paired. Anterior laterotergite subtriangular, one-half as large as subovate posterior laterotergite. Episternum bearing from 2 to 6 short spinelike setae. Eusternum without transverse row of setae. Legs rather short, strong. Coxa with up to 25 spinelike setae on anterior aspect, mostly in 2 oblique rows diverging mesally with conspicuous glabrous area between rows; 7 to 8 stout setae on poste- rior surface. Trochanter with 10 to 12 spinelike setae on medio- anterior surface, arranged in 2 rows; 6 to g such setae and 1 fine seta scattered on posterior surface ; 2 well-developed setae on medial aspect. Femur usually with 9 to 11 spinelike setae on medioanterior surface ; 4 to 6 spinelike setae and 1 slender seta on posterior surface; 1 long seta on medial aspect; 1 or 2 fine setae on lateral surface. Tibio- tarsus about as long as femur, but narrower ; with 5 or 6 setae around distal margin; 4 spinelike setae and 1 slender seta on medioanterior surface; 3 or 4 spinelike setae on posterior surface. Ungula, when uneroded, almost as long as tibiotarsus. 146 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III First to eighth abdominal segments subequal ; first segment short- est; fourth to sixth segments widest. Mediotergites (mtg, fig. 27, d) with few small, shallow pits; transverse branch of impression (trim) shallow, slightly sinuate, on second to eighth segments reaching from one-half to three-fourths distance from longitudinal branch to mid- dorsal suture, shorter on first segment; longitudinal branch of im- pression shallow, variable in length, extending from one-half to nearly total distance from transverse branch to posterior transverse row of setae. Anterior part of mediotergite without setae. Posteriorly on each mediotergite, transverse row of 5 setae on second to eighth seg- ments, only the most lateral seta being unpaired ; only 3 setae in poste- rior row on the first abdominal segment. Usually 1 small seta, rarely 2, along margin of mediotergite laterad to impression. Spiracles (sp, fig. 27, d) subequal in size, in anterior half of segment except in eighth abdominal segment where spiracle always in posterior half. Spiracular sclerite (spsc) well developed, subovate ; usually wider than spiracle and 2 to 3 times as long as spiracle, increasing in size from first to seventh segment, smaller in eighth segment. Laterotergite I (ltg I) extending length of segment; 1 large seta on dorsal margin. Pleurite large, subovate, with 1 large seta. Sternum of I piece, sub- quadrate ; 4 faint impressions, indefinite except on the more anterior segments ; with 1 stout seta at each corner. Ninth abdominal segment (fig. 27, c, d), exclusive of urogomphi, about as long as eighth abdominal segment and three-fourths as wide ; almost as long as wide; sides of anterior half subparallel, posterior half tapering caudally making width at anterior margin of caudal notch two-thirds (or less) greatest width of segment. Dorsum convex, more pronounced anteriorly. Dorsal plate (dpla) sloping downward from front to back, usually with a short transverse concavity slightly caudad to the 2 setae found near the middle of the plate; irregularly lined and wrinkled, but without pits; 4 faint longitudinal impressions, 2 laterally (lim) and a short paramedian pair (pim) in anterior part of plate ; lateral margins slightly raised and carinate, bearing 3 prominent blunt “teeth” (to), each with a long bristle; transverse impression (trim) feeble, usually interrupted in middle. Tergite (tg) continues uninterruptedly laterally and on posterior ventral surface; usually with from 5 to 10 unpaired setae on each side, some issuing from small sclerotized tubercles. Distance between posterior margin of pleural area (~/) and anterior margin of caudal notch from one-sixth to one-fifth of total length of segment (exclusive of urogomphi). Pleural area well developed, membranous, transversely striate. Ster- NO. II LARVAE OF THE ELATERID BEETLES—GLEN 147 num of 2 sclerites, separated anteriorly by median longitudinal suture, and posteriorly by tenth abdominal segment ; each sclerite with 2 or 3 setae in row around tenth abdominal segment. Urogomphi (ur, fig. 27, c, d) separate, robust, bifid; inner prong slightly longer and much more robust than outer prong. Inner prongs (ipr) short, stout, with hard, dark anteromedian margins; directed inwardly ; tips bluntly pointed, frequently meeting or overlapping ; each with 2 stout setae, I issuing from base of small tubercle on caudolateral margin and 1 from midventral surface. Outer prongs (opr) shorter and more slender than inner prongs, projecting upward, sometimes slightly outward and backward; each terminating in horny tip, sometimes with short, sharp point inclined inward ; 2 large setae, I on anteromedial surface of distal half of prong, the other at base of prong on lateroventral aspect. Undivided part of urogomphus short and very broad, with 1 stout seta ventrally just anterad to bases of prongs. Caudal notch (cv) small, subovate or shield-shaped, slightly longer than wide; very narrow posteriorly, often entirely closed by converg- ing inner prongs. Tenth abdominal segment (70) surrounded by Io prominent setae ; anal aperture linear and median. Material used in study.—Fourteen larvae were examined. These were collected from Denmark (6), Finland (5), and Germany (3). So far as is known, none of these specimens were collected at the same time and place as specimens that subsequently were reared, but the larva of this species has been known to European workers for many years and it is believed that the material is reliably named. The material examined is deposited in the Canadian national collection, the U. S. National Museum, and van Emden’s collection. * * * THE LUDIUS PYRRHOS GROUP FIGURES II, e; 28 KEY TO SPECIES Mrommeastecne NOLtheAmeni Canes cmc aictacieeeeier pytrhos (Herbst) (p. 148) From western North America................ protractus (LeConte) (p. 149) Knowledge of this group is limited to rather inadequate larval material of pyrrhos (Herbst) and protractus (LeConte). The group differs from all other known Ludius larvae in the character of the spiracles of the eighth abdominal segment (figs. 11, e; 28, a), which 148 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III are at least twice as long as the spiracles in the other abdominal seg- ments and situated much farther caudad. Other important diagnostic features are found in the nasale (fig. 28, c), which is tridentate at tip with small median denticle, and in the mandible (fig. 28, b) with its unusually robust distal half. The group is not closely allied to any other of the known Ludius larvae. Larvae of the pyrrhos group are yellowish brown, exceeding 20 mm. in length when full-grown. Dorsum punctulate. Caudal notch small, almost closed posteriorly. Urogomphi bifid, prongs subequal in length; inner prong broad; outer prong less robust, terminating in moderately blunt point. Dorsum of ninth abdominal segment (fig. 28, d) with 2 unpaired setae, and sometimes with faint short median groove; anterior margin of dorsal plate well defined throughout ; 3 rounded “‘teeth” (to), sometimes with pointed tips, on lateral mar- gins of dorsum. On ninth abdominal segment the distance between the pleural area and caudal notch equals one-quarter to one-third of entire length of segment, exclusive of urogomphi. Frontoclypeal area truncate posteriorly. One “sensory” appendix on second segment of antenna. Second joint of maxillary palpus longer than first joint. Eyes present, 2 prominent unpaired setae around each eye. Epicranial plates each with 2 prominent unpaired lateroepicranial setae. Gula short and very narrow. Presternum of prothorax of I piece, but deeply incised on lateroposterior aspects, almost resulting in a separate small median piece posteriorly. Without spinelike setae on episterna of mesothorax and metathorax. Mediotergites of mesothorax and metathorax with definite impressions. Mediotergites of abdominal segments (mtg, fig. 28, a) with transverse branches of impressions (trim) failing to reach middorsal suture and with a posterior trans- verse row of 5 prominent setae arranged as 2 pairs with 1 unpaired hair farther laterad. LUDIUS PYRRHOS (Herbst) FIGURES II, e; 28 Elater pyrrhos Hersst, Natursyst. Ins. Kafer, vol. 10, p. 30, 1806. Corymbites pyrrhos (Herbst), LeEContE, Trans. Amer. Philos. Soc., new ser., vol. 10, p. 447, 1853. Ludius pyrrhos (Herbst), VAN Dyke, Proc. California Acad. Sci., vol. 20, P. 394, 1932. The larval material of this species that was available for study was not adequate for a description of greater detail than that given under the pyrrhos group. NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 149 The third segment of the antenna is very short, apparently being less than one-half as long as the second segment. This might be a specific character. At present, the separation of the larvae of pyrrhos and protractus is best made on the basis of geographical distribution. Apparently the material examined was part of a collection made in a cornfield that was damaged by wireworms. This is the only infor- mation available on the biology of the species. Material used in study—According to the ledger records of wire- worm rearing at the United States Bureau of Entomology Laboratory, Hagerstown, Md., specimens given the accession numbers 6066 to 6080, inclusive, were collected together at College Park, Md., April 24, 1916, and were identified, probably by J. A. Hyslop, as Ludius pyrrhos (Herbst). Larval exuviae of 3 specimens, labeled 6066, 6067, and 6069, and a much desiccated whole larva, labeled 6075, were found and examined. The notes indicate that specimen No. 6068 was reared, and an adult of Ludius pyrrhos (Herbst) bearing this number has been reported by M. C. Lane to be in the collection at the Wireworm Laboratory, Walla Walla, Wash. Mr. Lane confirmed the identifica- tion of this adult. Unfortunately, the larval exuvium of No. 6068 was not available, but the evidence strongly suggests that the material that was examined was of this species. The larval material is now in the U. S. National Museum. LUDIUS PROTRACTUS (LeConte) Corymbites protractus LECoNTE, Proc. Acad. Nat. Sci. Philadelphia, p. 85, 1859. Ludius protractus (LeConte), VAN Dyke, Proc. California Acad. Sci., vol. 20, p. 410, 1932. Van Dyke (1932, pp. 392, 410) states that this species is widely distributed along the Pacific coast from British Columbia to northern California, and that it is a meadowland species. This suggests that the larva lives in the soil. The specimen examined was found as a pupa on June 19, which indicates a rather early seasonal development. Additional larval material is necessary before a complete detailed description can be given and before reliable structural characters are known for the separation of protractus and pyrrhos. Material used in study.—Only the larval exuvium of a reared speci- men was available for examination. W. J. Brown, of Ottawa, con- firmed the identification of the reared adult. 1; Towle District, Placer County, Calif.; June 19, 1932; collected as a pupa which emerged immediately; H. H. Keifer. (U.S.N.M.) x * * 150 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III THE LUDIUS LIMONIIFORMIS GROUP FIGURES II, h; 20 KEY TO SPECIES With 5 or more “teeth” laterally on dorsum of ninth abdominal segment (to, fig. 29, b) ; larva rarely exceeding 20 mm. in length; west of Lake Supebion. 2/2: + Moms crane eee mee limoniiformis (Horn) (?) (p. 151 With 4 “teeth” laterally on dorsum of ninth abdominal segment (fo, fig. 29, e); larva might exceed 30 mm. in length; eastern North America.... ices lehelnre caters ts SERRE OTe cylindriformis (Herbst) (?) (p. 156) This group is represented in the larval stage by two species that are not positively identified, but which are believed to be limontiformis (Horn) and cylindriformis (Herbst). These larvae are soil inhabiting, apparently preferring relatively moist situations. Both species are minor pests of cultivated plants. Ludius limonufornus and L. cylindriformis larvae are easily sepa- rated and might eventually be placed in different species groups, but they possess many characters in common and are not closely allied to any other of the known Ludius larvae. The body is relatively long and narrow, most of the abdominal segments being as long as wide. yes are absent and the presternum of prothorax is undivided. Other important characters are found in the urogomphi (fig. 29, b-e), the dorsum of the ninth abdominal segment (fig. 29, b, e)—especially the sharp marginal “teeth” (to) and the lack of a mediodorsal groove— and the nasale (fig. 29, a). The larvae of the limonuformis group are light brown or yellow brown. Caudal notch small, subcircular or subovate, nearly closed posteriorly. Urogomphi bifid; inner prong broad, with 1 or more tubercles on posterior aspect; tip of outer prong sharp, inclined in- ward or forward. Ninth abdominal segment (fig. 29, b, e) with 4 or more prominent, sharp “teeth” (#0) on lateral aspects of dorsum and with 4 unpaired setae on the central dorsal area, but without a median longitudinal groove. Nasale (fig. 29, a) with tridentate tip, lateral denticles small and frequently eroded off. Frontoclypeal area truncate posteriorly. Second joint of antenna bearing I “sensory” appendix. Basal joint of labial palpus without setae. Eyes absent. Gula short, narrow. Presternum of prothorax undivided. Episterna of meso- thorax and metathorax with several spinelike setae. Abdominal mediotergites with some of prominent setae paired (usually 2 pairs) or in a semipaired arrangement, and in second to fifth segments with transverse branches of impressions reaching to or nearly to the mid- dorsal suture. First 8 abdominal segments with small pleurites, NO. II LARVAE OF THE ELATERID BEETLES—GLEN I5I becoming narrower in more posterior segments; sternum undivided ; and spiracles in anterior half of each segment. LUDIUS LIMONIIFORMIS (Horn) (?) FIGURES I1, h; 20, a-c Corymbites limonuformis Horn, Trans. Amer. Ent. Soc., vol. 3, p. 320, 1871. Ludius limonuformis (Horn), VAN Dyke, Proc. California Acad. Sci., vol. 20, P. 420, 1932. According to Van Dyke (1932, pp. 396, 420) this species ‘“‘seems to be rather widely spread in the country to the west of the Great Lakes,” and apparently extending into Alberta. The larvae are known from the parklands of Saskatchewan, where they are found most frequently in association with the grassy margins of snowberry (Symphoricarpos) and silverberry (Elaeagnus) thickets. The species persists for many years after such areas are brought under cultivation. King (1928, p. 705) refers to L. limonufornis as a crop pest of minor importance in Saskatchewan. This larva resembles the more eastern cylindriformis (Herbst) but is smaller, and has more “teeth” on the margins of the dorsum of the ninth abdominal segment (fig. 29, b). Description of “mature” larva.—Length 16.5 mm. ; greatest breadth 2.0 mm.; thoracic and abdominal segments subequal in width with prothorax sometimes widest segment. Fully distended larva measured 22 mm. Body only moderately robust; with moderate-sized mem- branous areas on lateral aspect; most segments of abdomen as long as wide ; head and ninth abdominal segment about five-sixths greatest body width. Dorsum light brown to yellowish brown (near “‘clay color,” Ridgway, 1912) ; head and prothorax distinctly darker ; venter slightly paler. Dorsum slightly rugose, more pronounced rugosity on abdomen ; with sparse small punctures, more abundant on more poste- rior segments. Head subquadrangular with arcuate sides; flattened above and below. Frontoclypeal region with posterior part extending backward almost to foramen magnum; truncate posteriorly. Two prominent anterior nasosulcal setae on each side of base of nasale. Nasale (n, fig. 29, a) a well-developed tooth terminating tridentate when uneroded ; median denticle large, small lateral denticles frequently eroded giving appear- ance of unidentate tip. Subnasale (sm) consists of many fine, short, sharp, ‘forward-projecting denticles; frequently eroded.’ Paranasal 152 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III lobes produced beyond nasale ; each bearing 3 setae, I very small and sometimes lacking. Epicranial plates smooth, or with few small pits. Dorsal sulci shallow ; each with 5 setae subequally spaced, most anterior seta long, others small. Ventral sulci bearing row of 5 to 8 setae, usually only 2 or 3 conspicuous. Two large, unpaired, lateroepicranial setae. Eye spot absent; normal eye region bordered by 2 unpaired setae. Postgenal areas expanded mesally, almost meeting; glabrous. Gula short, narrow; goblet-shaped ; constricted posteriorly by con- verging postoccipital sutures; glabrous. Antenna with first joint weakly clavate; almost two-thirds as wide as long; without setae; 3 or 4 small pores. Second segment subcylindrical, about two-thirds as wide as long; three-fifths length of basal joint; I or 2 pores; a few small setae or pegs borne dis- tally; 1 medium-sized conical “sensory” appendix just ventrad to base of third joint. Terminal segment small, about one-half as long as second segment and one-quarter as wide; 4 setae and 2 or 3 “sensory” pegs on apex. Mandibles of moderate length, robust; about two-thirds as wide at base (ventral aspect) as long; retinaculum well developed ; peni- cillus sometimes reaching base of retinaculum. Distal half inward bending; pointed; outer surface convex with short dorsal groove; inner face slightly excavate with small median carina, ventral margin of inner face sharp and slightly convex ventrally, dorsal margin sharp and moderately convex dorsally. Ventral mouthparts about four-fifths as wide across bases of stipites as at anterior ends of stipites. Cardines slightly separated. Stipes large, subrectangular ; proxistipes and dististipes not distinct ; usually 4 or 5 prominent setae on antero-lateroventral aspect. Galea with basal joint slightly longer and one-half wider than terminal segment, without setae or pores; terminal joint curved, lateral surface longer than inner surface, 2 or 3 pores on lateroventral aspect. Maxillary palpi with all segments subcylindrical. First joint slightly longer than wide; distally on mesoventral aspect with group of about 6 small pores and 2 setae. Second segment as wide as basal joint and longer; without setae; 3 or 4 pores. Third segment about as wide as long; about one-half length of second joint; 2 pores ventrally ; distally with 1 minute seta on mesoventral aspect and 1 near lateral aspect. Fourth joint longer than wide; nearly as long as third joint; sometimes I minute seta on medial aspect; 1 pore on lateral surface. Postmentum with 1 long seta near each corner. First prementum NO. II LARVAE OF THE ELATERID BEETLES—GLEN 153 with 1 large seta just caudad to base of each palpus. Labial palpus with basal joint cylindrical, about one-half length of first prementum, about as long as wide, without setae, 3 or 4 pores; terminal joint shorter than basal joint and about one-half as wide, without setae, I or 2 pores. Prothorax about equal to combined length of mesothorax and meta- thorax ; wider posteriorly ; about as long as wide. Tergites minutely punctulate; anteriorly with 5 setae (on each side of median dorsal suture) in transverse row, arranged as 2 pairs with I unpaired seta between, the unpaired hair often found near the more lateral pair giving appearance of group of 3 setae; posteriorly with 4 setae in transverse row, arranged as I pair with 2 unpaired setae farther laterad; glabrous elsewhere. Episternum with 1 large and usually 2 smaller setae. Epimeron bearing 1 small seta. Presternal area con- sisting of 1 large triangular sclerite, posteriorly acute, striate on anterolateral aspects, I prominent seta near each lateral margin, and 2 short diagonal rows each of 4 minute setae anteriorly near middle. Eusternum small, membranous, with small narrow median sclerite. Sternellum and poststernellum indefinite, small and largely mem- branous. Mesothorax and metathorax each about twice as wide as long. Mediotergites sparsely punctulate; transverse branch of impression variable in length, commonly extending about one-fourth distance from longitudinal branch to middorsal suture; longitudinal branch of impression very short or absent. Anterior part of each medio- tergite sometimes with several minute setae issuing from punctures ; 2 or 3 small setae arranged as follows: 1 just behind transverse branch of impression, 1 just laterad to impression, and sometimes I near medial end of impression or slightly farther mediad. Pos- terior part of each mediotergite with transverse row of 4 conspicuous setae, the 2 most medial paired; sometimes I minute seta just caudad to the most lateral of the large hairs. Anterior laterotergite sub- triangular, one-half as large as subovate posterior laterotergite. Epis- ternum bearing up to 10 spinelike setae, usually 6 to 8. Mesothoracic spiracle subequal in size to spiracles in abdomen. Legs subequal in length. Coxa of prothoracic leg usually with from 25 to 32 spinelike setae on anterior surface, mesothoracic and metathoracic legs with from 20 to 27 such setae; 6 to 8 stout setae and a few finer hairs on posterior aspect. Trochanter with 5 to 8 spinelike setae on medioanterior surface; 5 to 7 such setae and I fine seta on posterior surface; 2 long setae on medial aspect. Femur 154 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III usually with 5 to 7 spinelike setae on medioanterior surface; 2 or 3 spinelike setae and 1 slender seta on posterior surface; I long seta on medial aspect; I or 2 minute setae on lateral surface. Tibiotarsus with 6 setae around distal margin; 2 or 3 spinelike setae on medio- anterior surface; 2 or 3 spinelike setae on posterior face. Ungula, when uneroded, almost as long as tibiotarsus. First to eighth abdominal segments subequal ; first segment short- est; seventh and eighth segments very slightly narrower. Medio- tergites somewhat rugose and with small shallow punctures, rugosity and punctation becoming stronger from first to eighth segments; transverse branches of impressions sinuate, reaching to or nearly to middorsal suture on second to fifth segments, sometimes nearly as long on sixth segment, shorter on other segments especially first and eighth; longitudinal branch of impression extending approximately one-half distance from transverse branch to posterior transverse row of setae. Anterior part of each mediotergite usually with 4 setae, 2 (rarely 3) lying within transverse branch of impression, I toward middle of sclerite, equidistant from branches of impression, and 1 laterad to impression. Posterior part of each mediotergite with 5 large setae arranged as 2 pairs (sometimes only semipaired) and 1 unpaired seta farther laterad; sometimes I minute seta near unpaired large hair; sometimes, especially on more posterior segments, up to 5 additional small setae, usually slightly posterior to row of larger setae. Laterotergite I extending length of segment; with 1 large seta. Spiracles subequal; in extreme posterior end of spiracular sclerite, which is small, subovate, about twice length of spiracle and situated in anterior half of each segment. Pleurite small, becoming narrower in more posterior segments; with I prominent seta and sometimes I or 2 minute hairs. Sternum of I piece, subquadrate; without defi- nite impressions or sutures, but with a few small shallow pits; bear- ing I prominent seta near each corner and usually a few minute setae along lateral margins. Ninth abdominal segment (fig. 29, b), exclusive of urogomphi, slightly shorter than eighth abdominal segment and almost as wide; as long as wide; sides of anterior half subparallel, posterior half tapering caudally making width at anterior margin of caudal notch about three-fifths to two-thirds greatest width of segment. Dorsum flat to slightly convex, sloping downward from front to back. Dorsal plate (dpla) roughened by irregular lines, wrinkles, and a few scat- tered pits; 4 faint longitudinal impressions, 2 laterally (lim) and a paramedial pair (pim) which converge posteriorly but do not meet; NO. II LARVAE OF THE ELATERID BEETLES—GLEN 155 2 prominent fine setae anteriorly and 2 smaller setae farther caudad ; lateral margins slightly raised and carinate, typically with 5 promi- nent, sharp, setiferous “teeth” (to), sometimes with smaller “teeth” between larger ones, as many as 8 or g “‘teeth” on each side in some specimens; transverse impression (trim) continues across segment. Tergite continues uninterruptedly laterally and on posterior ventral surface; usually with from 20 to 35 setae on each side, many issuing from small sclerotized tubercles; anteriorly on lateral aspect with a few pits. Distance between posterior margin of pleural area and anterior margin of caudal notch approximates one-fourth total length of segment (exclusive of urogomphi). Pleural area large, trans- versely striate with small ovate sclerite near anterior ends. Sternum of 2 sclerites, separated anteriorly by faint median suture and pos- teriorly by tenth abdominal segment; each sclerite with from 5 to 10 setae, mostly in irregular row adjacent to tenth abdominal segment. Urogomphi (ur, fig. 29, b; fig. 29, c) separate, bifid; projecting dorsocaudad ; prongs subequal in length, but differing in shape. Inner prong (ipr) robust, subquadrate (rarely subtriangular), projecting caudomediad ; anteromedial margin strongly sclerotized, dark; with I or 2 setiferous tubercles (tub) on caudolateral aspect, another seta just ventrad to tubercles and usually a few minute setae scattered over caudal aspect. Outer prong (opr) less robust, corniform, pro- jecting caudad or dorsocaudad, sometimes slightly laterad ; terminat- ing in sharp horny point usually curving slightly forward or inward; with a prominent setiferous tubercle situated caudolaterally near base of prong; I prominent seta on inner aspect, about halfway up prong, I smaller seta on posteromedial surface and I minute seta medially near base of prong. Undivided part of urogomphus short, broad; usually with 1 very short, fine seta near midventral aspect. Caudal notch (cz) small, subcircular, almost closed by incurving inner prongs. Tenth abdominal segment with whorl of Io fine setae, sometimes a few additional minute setae ; anal aperture linear and median. Material used in study.—Twenty-one larvae, all from Saskatoon, Saskatchewan, were examined. This species has not been reared, and identification is based upon field association of larvae and adults. However, misidentification is unlikely because nearly all other elaterid larval types occurring in fields at Saskatoon have been identi- fied through rearing. The material examined is deposited in the Canadian national collection. II 156 SMITHSONIAN MISCELLANECUS COLLECTIONS VOL. III LUDIUS CYLINDRIFORMIS (Herbst) (?) FIGURE 20, d, e Elater cylindriformis Hersst, Natursyst. Ins. Kafer, vol. 10, p. 93, 1806. Corymbites cylindriformis (Herbst), GerMar, Zeitschr. fiir die Ent., vol. 4, p. 64, 1843. Ludius cylindriformis (Herbst), VAN Dyxe, Proc. California Acad. Sci., vol. 20, p. 399, 1932. This species is known from the eastern provinces of Canada and from the northeastern United States and at least as far west as In- diana, where Blatchley (1910, p. 765) reports its occurrence as “frequent.” The larvae have been taken from cornfields and gardens, and Hyslop (1915a, p. 9) reports the adults as being abundant in al- falfa and wheat fields. Hawkins (1936, pp. 55-56) implies that the species is a minor crop pest in Maine, but a predaceous tendency is suggested from the observations of Devereux (1878, p. 143), who found a larva crushing the elytron of a living Harpalus pennsyl- vanicus. A hibernating larva was found on December 15 at a depth of 1 foot in a cornfield.’? No adequate description of the larva has been published, but Haw- kins (1936, p. 59, fig. A) figured the ninth abdominal segment. Dis- tinction from the larva of limoniiformis (Horn) is made primarily through greater size, characters of the ninth abdominal segment (fig. 29, e), and the mandibles. The largest larva examined was not fully distended, but measured 30 mm. in length and 2.6 mm. in breadth. The mandibles are dis- tinguished as follows: Beginning at the distal end of the antennal fossa, a shallow groove extends one-third to one-half the distance toward the tip of the mandible. The ninth abdominal segment (fig. 29, e) bears only 4 sharp “teeth” on each lateral margin of the dorsum; distance between pleural area and caudal notch approxi- mates one-fifth to one-sixth total length of segment, exclusive of urogomphi ; lateral and ventral aspects of tergite with fewer setae than limontiformis, usually less than 20 setae on each side. The larvae examined possessed a few transverse rugae or pits on the abdominal mediotergites, especially within the angle of the impressions. Material used in study.—Five larvae were examined, four from Springhill, New Brunswick, and one from Riverton, N. J. The species has not been reared, but the material examined appears to be similar to the larva described by Hawkins (1936) as L. cylindri- 12 This observation was made at Riverton, N. J., by C. A. Thomas, of Kennett Square, Pa., and the larva was identified by the writer. NO. II LARVAE OF THE ELATERID BEETLES—GLEN M7 formis. Specimens examined are deposited in the Canadian national collection and the Pennsylvania Agricultural Experiment Station collection. Genus LIMONIUS Eschscholtz FIGURES 9, 9g; 10, e; II, a, d, g; 30-32 In the present study, the name Limonius Eschscholtz is used in the broad sense, including all species that have been placed in Pheletes Kiesenwetter and Nothodes LeConte. The larval morphology sug- gests this procedure to be best for the present at least. Hyslop (1921) has shown Pheletes to be isogenotypic with Limonius, and Van Dyke (1932, p. 333) suppresses the name Nothodes, placing its lone species, dubitans LeConte, in Limonius. Accurate specific identification of the larvae of this genus is made doubly difficult because of the apparent unsettled status of the tax- onomy of the adults. A thorough revision of the genus would greatly assist in naming the larvae. Larvae of 14 species of Limonius have been studied, 12 from North America and 2 from Europe. Dr. Van Dyke (1932, pp. 333, 360) states that with one excep- tion the genus is holarctic in distribution and shows a preference for valleys and more open places. The great majority of the larvae studied are soil inhabiting and in North America are important pests in fields and gardens. Only one species, unidentified, is known to occur in decaying wood, but a few prefer moist forest litter. On the basis of rather scanty information it appears that pupa- tion occurs most commonly in July or early August. The newly developed adults pass the winter in their pupal chambers. Limonius is part of that large, complex group whose larvae possess an undivided prosternum (prst, fig. 31, c), have impressions on the mediotergites of mesothorax and metathorax, and with one excep- tion, L. pilosus (Leske), have the nasale tridentate at the tip. On the basis of larval characters the closest allies appear to be Elathous bicolor (LeConte), Ludius resplendens (Eschscholtz), and species of the Ludius limoniuformis group. Generic separation is greatly facilitated if Limonius larvae are considered in two artificial groups: Group I, caudal notch small; outer urogomphal prongs reduced to the size of small tubercles (figs. 30, d, f; 31, d, e¢). Group IJ, caudal notch small; outer uro- gomphal prongs at least moderately well developed (figs. 30, h; 31, f; 32, d, e), sometimes as long as inner prongs; and ninth abdominal segment (figs. 30, h; 32, d) without a mediodorsal groove and with well-rounded “‘teeth” (to) on lateral aspects of dorsum. Group I is 158 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III distinct from all other known larvae in this complex of related genera. Group II is separated from Elathous bicolor and Ludius resplendens through the absence of the mediodorsal groove on the ninth abdominal segment, and from larvae of the Ludius limoniuformis group by the blunt “teeth”? on the same segment. Limonius larvae may be characterized as follows: Dorsum pale yellow (or orange-yellow) to yellow-brown, with minute (incon- spicuous) to moderate-sized punctures, but lacking prominent trans- verse rugae. Caudal notch small. Urogomphi bifid, prongs subequal or outer prongs shorter than inner prongs. Ninth abdominal seg- ment usually without a mediodorsal groove, and without “teeth’’ or with well-rounded “teeth” on lateral margins of dorsum. Nasale vari- able, but usually tridentate at tip. Frontoclypeal area truncate or broadly rounded posteriorly. Eyes present or absent, with 2 to 4 setae around eye region. Usually 2 unpaired lateroepicranial setae on each gena. Gula short and narrow. One “sensory” appendix on second segment of antenna. Mandibles variable (figs. 30, c, e; 32, 0). Presternum of prothorax (fig. 31, c) undivided. Mesothorax and metathorax with impressions on mediotergites. Abdominal medio- tergites with impressions of varying length; and usually with 5 to 7 prominent setae in posterior transverse row, definitely arranged as 2 or 3 pairs in nearly all species. KEY TO “SPECIES GROUPS” AND ISOLATED SPECIES OF LIMONIUS 1. Outer prongs of urogomphi reduced to the size of small tubercles (opr, SOFC OFT pet pene TAL o/h) | i EN oR aM RIAN Ai SIG Cass odode ooc 2 Outer prongs of urogomphi pronglike (opr, figs. 30, h; 31, f; 32,d,e).. 5 2. Ninth abdominal segment with 2 conspicuous conical protuberances (pro, AL Rape PEs AaC0 ee ae eten Reeet Cty acre RT TCH ET Sy OAC pectoralis LeConte (p. 161) Ninth abdominal segment without such protuberances.................. g) 3. Mandible (fig. 30, c) with prominent toothlike expansions anterior to retinaculum; nasale (n, fig. 30, a, b) short, with 3 subequal teeth.... Ay Ree RA SLRs tan Wears Mo ARS on gE the aeneoniger group (p. 159) Mandible without toothlike expansions anterior to retinaculum.......... 4 4. Eyes absent; ninth abdominal segment without a definite mediodorsal groove; each paranasal lobe with a cluster of 12 or more “sensory” POTes (Pulp, Me. 4200) s eae pee Nyala vn aeger LeConte (p. 163) Eyes present ; ninth abdominal segment with a mediodorsal groove; para- nasal lobes without definite cluster of pores...:....5: 1.006 «meee Ee EL Re unidentified, possibly confusus LeConte (p. 162) 5. Nasale (n, fig. 30, g) of 1 pointed tooth; mandible (fig. 30, e) with very large retinaculum and expanded medioventral cutting edge; Buropeaniie: os acces Hoek ....pilosus (Leske) (?) (p. 160) Nasale (n, figs: 31, b; 32, a) with tip tridentate; mandible (fig. 32, b) without above characteristics ; North American. .the canus group (p. 164) NOPE LARVAE OF THE ELATERID BEETLES—GLEN 159 THE LIMONIUS AENEONIGER GROUP FIGURES 9, 9; II, a; 30, a-d, f Only two species are included in this group, one from Europe, the other from western North America. The larva of the genotype, Limonius aeneoniger (DeGeer) (= Pheletes Bructeri Panzer), is recorded by Beling (1884, p. 205) from earth in dry, sunny places in European forests, preferably under moss. H. P. Lanchester, of Walla Walla, Wash., collected larvae of the Pacific coast species, L. consimilis Walker, “in moss on boulders in shade.” This strongly characterized group is distinguished as follows: Mandible (fig. 30, c, probably considerably eroded) with prominent toothlike expansions anterad to retinaculum; nasale (n, fig. 30, a, b) short, with 3 subequal teeth; subnasale (sn) with 1 prominent obtuse tooth ; and abdominal mediotergites with transverse branches of im- pressions short, extending less than one-half distance from longi- tudinal branches to middorsal suture. Preserved larvae are yellow to yellow-brown in color, with head and prothorax darker, but living specimens appear to differ since Beling (1884, p. 205) reports “larva intense yellow-red” and Lan- chester (1941, p. 368) records consimilis as “orange-yellow” when collected. Both are relatively small species, aeneoniger attaining 9 mm. in length, and the largest consimilis examined measuring 13 mm. Urgomphi (ur, fig. 30, f; fig. 30, d) with outer prongs (opr) very small. Ninth abdominal segment (fig. 30, f) without mediodorsal groove (sometimes with a shallow, central, subcircular depression) ; without setae on central dorsal area; “‘teeth’’ (to) very small, prac- tically wanting, on lateral aspects of dorsum; and distance between caudal notch and pleural area about one-fifth length of segment, exclusive of urogomphi. Frontoclypeal area (fig. 30, a) truncate or broadly rounded posteriorly. Paranasal lobes (pul) without definite pore cluster. Eyes present. Two prominent unpaired lateroepicranial setae on each gena. Without setae on ventral aspect of basal segment of labial palpus. Mesothorax and metathorax without spinelike setae on episterna (usually 1 fine seta on each episternum). Abdominal mediotergites finely punctured and each with 2 or 3 pairs of setae in posterior transverse row. Material used in study.—L. aeneoniger (DeGeer): Only 2 speci- mens were available for examination, 1 from Norway, the other from Denmark. As far as is known, neither of these was associated 160 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Eid with reared material. Specimens deposited in the U. 5S. National Museum. L. consimilis Walker: Nineteen specimens collected at Asotin, Wash., were examined. These were received through the cour- tesy of M. C. Lane and H. P. Lanchester, of Walla Walla, Wash. Mr. Lanchester reared larvae from the same collection. The reared adults were identified by M. C. Lane and placed in the collection of the Wireworm Laboratory at Walla Walla. Larvae examined are deposited in the Canadian national collection. LIMONIUS PILOSUS (Leske) (?) FIGURE 30, e, g, h The larva of this European species has been described by Beling (1883, p. 302; 1884, p. 205), Rey (1887), and Henriksen (1911, pp. 255-256) under the name of nigripes Gyllenhal. In the present study examination has been confined to one larva from Amager Falled, Denmark, that is believed to be identical with that described by these authors. This specimen is deposited in the U. S. National Museum. According to Beling and Henriksen, the larva inhabits the soil, chiefly in meadowland. This larva differs from all other Limonius larvae known to the writer, as follows: Nasale (n, fig. 30, g) unidentate, sharply pointed when uneroded; subnasale (sv) consisting of a transverse ridge bear- ing about 8 subequal forward-projecting denticles; and mandible (fig. 30, e) with greatly enlarged retinaculum (ret) and expanded ventral cutting edge. Rey (1887) states that the larva of Limonius cylindricus Paykull is very similar. Length 14.5 mm., as recorded by Henriksen and Beling. Outer prongs of urogomphi (ofr, fig. 30, 1) shorter than inner prongs (7pr), but definitely pronglike. Dorsum of ninth abdominal segment (fig. 30, ) apparently without a median sulcus, without setae on central area, and with small, blunt “teeth” (to) on lateral aspects. Fronto- clypeal area truncate posteriorly. Eyes present. Two unpaired lateroepicranial setae on each gena. Five or six spinelike setae on each episternum of mesothorax and metathorax. Abdominal medio- tergites with small punctures, prominent setae arranged in definite pairs, and with transverse branches of impressions extending about four-fifths of distance from longitudinal branches to middorsal suture. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 161 LIMONIUS PECTORALIS LeConte FIGURE 31, a, ¢, 9 This northern species is known from Alaska and from coast to coast in Canada. Larvae have been reared from Cleeves, Saskatche- wan. This species is of considerable economic importance in the park- lands of northern Saskatchewan and the Peace River Block of Alberta and British Columbia. As reported by King (1928, p. 704), pecto- ralis appears to be definitely associated with deep, rich, loam soils, and somewhat low-lying situations, and under native conditions prefers the grassy margins of snowberry (Symphoricarpos) thickets. After a lengthy period of drought larvae were found on June 30 as deep as 19 inches in an old bromegrass field. However, it would appear that pupation occurs much nearer the soil surface, since over- wintering adults have been taken at depths between 2 and 4 inches. Rearing records indicate that pupation occurs in late summer or early autumn, but the exact time is not known. The larva of pectoralis differs from other known elaterids in having 2 prominent conical protuberances on the dorsum of the ninth abdomi- nal segment (pro, fig. 31, e, g). The nasale (fig. 31, a) is also char- acteristic, the median denticle being smaller than the lateral denticles. The largest larvae examined measured 14 mm. Outer prongs of urogomphi (opr, fig. 31, e, g) are reduced to mere, pointed tubercles. Ninth abdominal segment without a mediodorsal sulcus, without “teeth” (rarely I or 2 minute swellings) on lateral aspects of dorsum, and distance between caudal notch and pleural area approximates one- eighth to one-sixth length of segment, exclusive of urogomphi. Frontoclypeal area truncate posteriorly. Eyes absent. Two unpaired lateroepicranial setae on each gena. Mandible of common lepturoidine type, with well-developed retinaculum. One small seta on ventral sur- face of basal segment of labial palpus. Mesothorax and metathorax with impressions indistinct on mediotergites, and with 4 to 6 spinelike setae on each episternum. Abdominal mediotergites with impressions reaching the middorsal suture on second to eighth segments, with fine, scattered punctures, and with 3 pairs of setae in posterior trans- verse row. Material used in study.—Fifteen examples were examined, includ- ing the larval exuviae of two reared specimens. All the material was from the parklands of Saskatchewan except 7 larvae from Fair- banks, Alaska. The following notes pertain to the reared material : 2; Cleeves, Saskatchewan; (no date) ; both reared to adults by Nov. 15, 1926; I. M. King. (C. N. GC.) 162 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III LIMONIUS CONFUSUS LeConte (?) A few larvae of this species have been collected at Urbana and Dan- ville, Ill., and at Arnprior, Ontario. This is a typical forest species, the larvae occurring in decaying wood and in moist, well-decayed leaf litter. The identity of these larvae is unknown, but it is most likely to be either L. confusus LeConte or a closely allied species. W. J. Brown, of Ottawa, informs the writer that confusus LeConte is found in the woods and is the most common Limonius at Arnprior. Blatchley (1910, p. 758) reports that the adults of confusus are found beneath the bark of oak and other trees, but that the species is scarce in Indiana. It is possible that there is some error in the reference by J. J. Davis (1911, p. 251) to L. confusus larvae damaging truck crops in Illinois. I have examined the larval exuvium of a specimen collected by Mr. Davis, the reared adult of which was identified by J. A. Hyslop as Limonius confusus LeConte. This larval skin bears structures of great similarity to those of the larva of L. dubitans LeConte, and on this basis Mr. Davis’ species would be placed in the canus group, which includes L. canus and L. dubitans and their close allies. However, Mr. Brown, of Ottawa, informs me that, on the basis of adult char- acters, confusus and dubitans belong in distinct species groups. Blatchley (1910, p. 762) states that L. dubitans ‘“‘resembles so closely Limonius griseus that it can with difficulty be separated.” Both of these species occur commonly in Indiana and undoubtedly in Illinois. Davis’ larva is more likely to be one of these soil-inhabiting species than the woodland confusus. It is unfortunate that the adults reared by Mr. Davis are either lost or misplaced and Mr. Hyslop’s original identification cannot be checked. The larva which is provisionally named confusus LeConte in the present study belongs to that group of Limonius that have very small outer urogomphal prongs. Its other chief characteristics are the presence of a mediodorsal groove on the ninth abdominal segment ; prominent setae unpaired on abdominal mediotergites ; eyes present ; and paranasal lobes without a definite cluster of pores. The largest larva examined measured 18 mm. in length. The ninth abdominal segment bears 2 very small (inconspicuous) setae anteriorly on dorsal plate; small, blunt “teeth” on lateral aspect of dorsum; and distance between caudal notch and pleural area approximates one-fifth total length of segment, exclusive of urogomphi. Nasale with tip NO. II LARVAE OF THE ELATERID, BEETLES—GLEN 163 tridentate. Frontoclypeal area truncate posteriorly. Three latero- epicranial setae on each gena, arranged as a dorsal pair and I unpaired hair farther ventrad. Mandibles robust. Without setae on ventral surface of basal segment of labial palpus. With 1 to 4 setae (rarely spinelike) on each episternum of mesothorax and metathorax. Abdominal mediotergites with impressions not extending to middorsal suture ; and bearing sparse, small punctures. Specimens examined are deposited in the Canadian national col- lection. LIMONIUS AEGER LeConte FIcurRES II, d; 31, d According to Van Dyke (1932, p. 339) this species occurs from Nova Scotia and the New England States west to British Columbia and south through the Rocky Mountains to New Mexico and from the Cascade Mountains to Mount Hood, Oreg. Larvae have been reared from Saskatoon, Saskatchewan. Under Saskatchewan conditions this species is found in forest litter and in the superficial layers of soil under the litter. The larvae have been collected in greatest abundance under mixed shrubs and poplar along the banks of the Saskatchewan River. In the muck soils of southern Quebec and Ontario, larvae believed to be of this species are injurious to vegetables. Rearing records indicate that pupation nor- mally occurs in soil or litter from late July to early August. The larva of aeger belongs to that group of Limonius that have very small outer urogomphal prongs (opr, fig. 31, d). It shows super- ficial resemblance to the larva provisionally regarded as confusus LeConte, but differs as follows: Eyes absent; ninth abdominal seg- ment without mediodorsal groove (sometimes with a shallow subcircu- lar depression bearing short, paired paramedian grooves), and with a sinuate “impression” laterally near base of segment; prominent setae paired on mediotergites of abdominal segments; each paranasal lobe bearing a cluster of 12 or more “sensory” pores; and usually 4 to 7 spinelike setae on each episternum of mesothorax and metathorax. The largest larvae attain 12 mm. in length. Ninth abdominal seg- ment without setae on central dorsal area, small and much-rounded “teeth” on lateral aspect of dorsum, and distance between caudal notch and pleural area approximating one-fifth to one-fourth length of seg- ment, exclusive of urogomphi. Nasale with tip tridentate. Subnasale of many fine teeth. Frontoclypeal area truncate posteriorly. Two prominent unpaired lateroepicranial setae on each gena. Abdominal 104 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III mediotergites with fine scattered punctures, and with transverse branches of impressions reaching about four-fifths of the distance from the longitudinal branches to the middorsal suture. Material used in study.—Thirteen examples were examined, in- cluding the larval exuviae of three reared specimens. All material was from Saskatoon, Saskatchewan, and is now in the Canadian national collection. The reared adults were identified by W. J. Brown, of Ottawa. Notes follow on the reared material. 6; Saskatoon, Saskatchewan; July 13, 1935; 1 reared to adult July 30, 1935; R. Glen. 7; Saskatoon, Saskatchewan; July 21, 1937; 2 reared to adults Aug. 3, and Aug. 20; R. Glen and H. McDonald. THE LIMONIUS CANUS GROUP IGURES 10; @5\ 11,9531, 0,.6, fs 32 PROVISIONAL KEY TO SPECIES 7° 1. From eastern North America, including the Great Lakes region....... 2 From western North America, including the Great Plains region....... 3 to Abdominal mediotergites (fig. 32, c) with impressions extending to or nearly to the middorsal suture on second to fifth segments; New Work: towlndianat cy cotton aan dubitans LeConte Abdominal mediotergites with shorter impressions; eastern Canada and {ectypus (Say) (?) (= northeastern United States.....:........ agonus Say) anceps LeConte (?) 3. Urogomphi with outer prongs hook-shaped (opr, fig. 31, f; eastern Washington, and Idaho to northern California. .subauratus LeConte (?) Outer urogomphal prongs not hook-shaped (fig. 32, ¢, f).......eseeeeee 4 4. Abdominal mediotergites with impressions (as in fig. 32, c) extending to or nearly to the middorsal suture on second to fifth segments...... 5 Abdominal mediotergites with shorter impressions; British Columbia and thesPacific Statesss a2, cea non bees Doe ere eee canus LeConte 5.) Hrome Manitoba, Saskatchewan, and Alberta) erie ete creer eerie Be re te a a aC Limonius sp., near ectypus (Say) f californicus (Mannerheim) \ occidentalis Candeze (?) The larvae of eight North American species have been included in this group. Of these, L. canus LeConte and L. dubitans LeConte 13 Since this study was completed, H. P. Lanchester has published a paper containing a key to the larvae of six species of this group: “Larval Determina- tion of Six Economic Species of Limonius,’ Ann. Ent. Soc. Amer., vol. 39, pp. 619-626, 1946. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 165 are believed to be reliably named. The larvae regarded as L. ectypus (Say) and L. anceps LeConte have been identified provisionally on the basis of geographic distribution and field association with adults of these species. Considerable uncertainty surrounds the identification of the larvae here designated as subauratus LeConte and occidentalis Candeze. The following changes might be in order: L. subauratus changed to infuscatus Motschulsky and occidentalis changed to sub- auratus LeConte. However, the names used in the present discussion are those that were found on the labels accompanying the specimens examined. The canus group includes some of the most important pest species in America. All are soil inhabiting and all have been reported as injur- ing cultivated plants. With reference to the Pacific coast species, Lane (1935, p. 530) reports that the larvae inhabit moist soils, such as stream margins, low-lying lands, and irrigated fields, and thrive espe- cially in alkaline areas. Similar habits have been observed for the larvae of “Limonius sp. near ectypus (Say)” occurring in the Prairie Provinces. The eastern species also are known to inhabit moist areas, but frequently are taken from sandy soils that are relatively well drained. Very little has been published on the life history of these species, but it appears that pupation normally occurs in late July or early August, with the adults overwintering in their pupal chambers. The morphology of the larva of L. canus LeConte has been studied in detail by Lanchester (1939). Larvae of the canus group differ from other known Limonius larvae in that the prongs of the urogomphi are subequal in length. This group is further characterized as follows: Relatively large larvae, sometimes attaining 25 mm. in length when mature; each paranasal lobe with a cluster of 12 or more “sensory” pores (pnp, fig. 32, a) ; eyes absent; with 1 small seta on ventral aspect of basal segment of labial palpus; maxillary palpus with second segment longest ; 6 to 12 spinelike setae on each episternum of mesothorax and metathorax ; abdominal mediotergites with prominent posterior setae arranged in pairs (fig. 32, c) ; dorsal plate of ninth abdominal seg- ment (figs. 11, g; 32, d, f) with 2 or 4 unpaired setae, without medio- dorsal groove (sometimes with shallow subcircular depression), and with well-rounded “teeth” (to) or tubercles on lateral margins; in ninth abdominal segment the distance between caudal notch and pleural area approximates one-sixth total length of segment, exclusive of uro- gomphi. 166 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Material used in study of the group.—L. anceps LeConte: Exami- nation was restricted to 10 larvae collected at Maxfield, Maine. These were provisionally identified by Dr. A. G. Boving, Washington, D. C., as Limonius anceps LeConte. (U.S.N.M.) L. californicus (Mannerheim): Only four larvae were examined. These were reared from eggs secured from adults collected at Walla Walla, Wash., by M. C. Lane, and identified by him. (U.S.N.M.) These larvae were found to be identical to the larvae of L. canus LeConte, an error in labeling being the most plausible explanation. Consequently, the key characteristics of L. californicus larvae were taken from Lanchester (see footnote, p. 164). L. canus LeConte: Examination was restricted to four larvae reared from eggs secured from adults collected at Walla Walla, Wash., by M. C. Lane and identified by him. (U.S.N.M.) L. dubitans LeConte: Six larvae were studied. These were col- lected from Cornwells, Bucks County, Pa., by C. A. Thomas. Mr. Thomas has reared this larval type and he identified the larval material used as L. agonus Say. In a personal communication, Mr. Thomas informed the writer that the species previously identified as L. agonus Say is now regarded as being L. dubitans LeConte. (U.S.N.M.) L. ectypus (Say) (?): Ten larvae collected at Chatham, Ontario, were examined. These specimens were taken from cultivated fields where adults of this species were very abundant, but identification was not confirmed by rearing. (C.N.C.) Limonius sp., near ectypus (Say): Twelve examples, including the larval exuvium of one reared specimen, were studied. These speci- mens were collected from Souris, Manitoba (9g), Taber, Alberta (2), and Radisson, Saskatchewan (1). W. J. Brown identified the reared adult. Material associated through rearing is listed below. 6; Souris, Manitoba; June 8, 1938; 1 reared to adult Aug. 16, 1938; H. W. Moore. (C.N.C.) L. occidentalis Candeze (?): Three larvae were examined. These were in the same vial as five adults, all from Toppenish, Wash. As far as is known the identification was made on the basis of field asso- ciation of larvae and adults. (U.S.N.M.) L. subauratus LeConte: Four larvae from Spokane, Wash., were studied. These were taken from the Hagerstown, Md., collection and were labeled, “Web. No. 4675.” Ledger records stated that two larvae bearing this number had been reared and their adults identified by J. A. Hyslop as Pheletes subauratus LeConte. However, neither the larval exuviae nor the reared adults could be found. (U.S.N.M.) NO. II LARVAE OF THE ELATERID BEETLES—GLEN 167 Genus ELATHOUS Reitter ELATHOUS BICOLOR (LeConte) FIGURES II, C; 33, a, b Elathous bicolor (LeConte) is the only species of this genus known in the larval stage. The larvae were collected from decaying wood at Baldwin Hill, Douglas County, Kans., on January 3. One specimen pupated on May 15 and the adult emerged June 4. This species is undoubtedly closely related to Athous and Lep- turoides, from which it differs by the combined characters of a small caudal notch and the outer urogomphal prongs being not longer than the inner prongs. It bears resemblance to the larva of Ludius resplendens (Eschscholtz), differing by having the outer urogomphal prongs sharp and inclining forward, and by the impressions on the abdominal mediotergites failing to reach the middorsal line. Distinc- tion from Limonius larvae is obtained through the combination of prominent outer urogomphal prongs and the presence of a dorsal median groove on the ninth abdominal segment. The principal characters of the larva of E. bicolor may be sum- marized as follows: Length 14 mm., not fully distended and probably not mature. Dorsum yellow brown. Caudal notch small. Urogomphi (ur, fig. 33, 0; fig. 33, a) bifid; prongs subequal in length or outer prongs slightly shorter than inner prongs; outer prong (opr) corni- form, projecting dorsad with sharp tip inclined forward. Ninth abdominal segment (fig. 33, 6) with median dorsal groove (mg), without setae on central dorsal area, with 3 well-rounded “teeth” on each side of dorsum, and with distance between pleural area and caudal notch about one-fifth total length of segment, exclusive of urogomphi. Nasale with tridentate tip. Eyes present; surrounded by 3 or 4 unpaired setae. Two prominent lateroepicranial setae on each gena, the more dorsal seta paired with a very minute seta. Gula short and narrow. One “sensory” appendix on second segment of antenna. Mandible with well-developed retinaculum. Presternum of prothorax undivided. Mesothorax and metathorax with definite impressions on mediotergites and 3 or 4 spinelike setae on each episternum. Abdomi- nal mediotergites with transverse branches of impressions reaching about three-fourths of distance from longitudinal branches to mid- dorsal suture, and bearing small to moderate-sized shallow pits, but lacking transverse rugae. Material used in study—Two examples were examined, one being the exuvium of a reared specimen, the adult of which was identified 168 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III by W. J. Brown, of Ottawa. Both specimens were collected at the same time and in the same woods, but from separate decayed logs. 2; Baldwin Hill, Douglas County, Kans.; Jan. 3, 1933; I reared to an adult June 4, 1933; C. H. Hoffman. (C.N.C.) Genus LEPTUROIDES Herbst FIGURES 12, b; 33, c-e In following Hyslop (1921), the generic name Lepturoides Herbst is used in preference to Campylus Fischer and Denticollis Piller and Mitterspacher, which are still commonly used by European workers. Larvae of this genus are known for the European genotype, Lep- turoides linearis (Linnaeus), and for the North American denticornis (Kirby), and a closely allied but unidentified larva from Alaska, which possibly is fulvus (Motschulsky). These larvae are very simi- lar and further study and more material are required to secure reliable separating structural characters. All known Lepturoides larvae inhabit the forest, usually occurring under the bark of decaying wood, especially in stumps and logs of deciduous trees. They are believed to be predaceous. Henriksen (1911, p. 269) states that L. linearis pupates in May and June, the newly formed adults emerging at once from their pupal chambers. In Pennsylvania, Knull (1934, p. 208) found a pupa of Lepturoides productus (Randall) on April 24, from which the adult emerged on May 4; and the author collected larvae of L. denticornis in Minne- sota on April 8 from which two adults developed by May 4. From these observations, it would appear that in this genus hibernation is limited to the larval stage, that the overwintering larvae normally pupate with the first warm weather, and, in contrast to most Lep- turoidini, the newly developed adults immediately leave their pupal cells for mating and egg laying. On the basis of larval characters, Lepturoides is closely related to Athous. It is readily separated by the following combination of char- acters: Dorsum dark brown to black-brown; caudal notch small; outer urogomphal prongs (ofr, fig. 33, d, e) very long, curving upward and forward; inner prongs (ipr) short and smooth, without posterior tubercles ; abdominal mediotergites (fig. 33, c) punctulate, but with- out transverse rugae. The larvae may exceed 20 mm. in length. Dorsum usually very dark, sometimes with a reddish undertone, darkest on prothorax and head ; venter pale yellow or cream-colored. Ninth abdominal segment (fig. 33, d) with median dorsal groove (mg), without setae on cen- NO. II LARVAE OF THE ELATERID BEETLES—GLEN 169 tral area, with 3 or 4 prominent, sharp “teeth” (to) on each lateral margin, and with distance between pleural area and caudal notch about one-fifth total length of segment, exclusive of urogomphi. Nasale tridentate at tip. Subnasale with many fine teeth. Fronto- clypeal area truncate posteriorly. Eyes large ; bordered by 2 prominent unpaired setae. Two large unpaired lateroepicranial setae on each gena. Gula short and narrow. One “sensory” appendix on second segment of antenna. Mandible with prominent retinaculum, but with- out other teeth or toothlike expansions. Without setae ventrally on basal segment of labial palpus. Presternum of prothorax undivided. Mesothorax and metathorax with definite impressions on medio- tergites, and up to 7 (usually 3 or 4) spinelike setae on each epister- num. Abdominal mediotergites (fig. 33, c) with impressions reach- ing mediodorsal suture on all or most of segments, and with small fine punctures, but lacking transverse rugae. Henriksen (1911, p. 269, fig. 62) depicts the larva of L. linearis with prominent rugae on the abdominal mediotergites. This condition was not found on any of the material examined. PROVISIONAL KEY TO SPECIES OF LEPTUROIDES Tee EGOMPe Ms ULOPemae areca ie He semis aioic eit icine ts crete srenrers linearis (Linnaeus) Brome Northe Americas se eaacccen cs selects obs hoes siecle sin 8 wismela tele 2 From eastern and central United States and Canada; impressions usu- ally reaching middorsal suture on all abdominal mediotergites (fig. to BG ce Ch lwlse, e ciatind 'a, SrehelS a-alntc oa via Ais atte ea chet rs wis) gaaeis denticornis (Kirby) From Alaska; impressions usually not reaching middorsal suture on first and eighth abdominal segments............ fulvus (Motschulsky) (?) Material used in the study of the genus—L. linearis (Linnaeus) : Thirteen larvae were examined. These were from Denmark (7), Germany (1), Finland (1), and England (4). None of the specimens was known to be associated with reared material. Labeled larvae were received from the U. S. National Museum and from the British Museum of Natural History. These specimens undoubtedly were of the same species and definitely were congeneric with reared larvae of Lepturoides denticornis (Kirby). Therefore, it is believed that the material was properly identified. L. denticornis (Kirby): Eight examples of this species were stud- ied, including the exuviae of two reared specimens. All specimens were taken together from under the bark of an old fallen elm. W. J. Brown, of Ottawa, identified the reared adults. 8; Carter County, Minn.; Apr. 8, 1933; 2 reared to adults on May 4, 1933; RAGleny .¢G2N-€)) 170 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Lepturoides fuluus (Motschulsky) (?): Only one larva was ex- amined. This specimen was from Alaska and the identification was made entirely on the basis of locality. (U.S.N.M.) Genus ATHOUS Eschscholtz 14 FIGURES 12, a, C; 34-36 Larvae of 14 species of Athous have been studied. These are pri- marily forest forms, living in the litter and decaying wood, and probably are chiefly predaceous. A few European species are found in woodland meadows and cause injury to plants when such areas are brought under cultivation. Larvae of this genus have not been recorded as crop pests in North America. In America, Athous larvae collected in May and June frequently have pupated within a few days after being brought indoors. These findings suggest that some species of Athous normally transform to adults in June or early July. However, with species of the vittatus group, pupation is reported to occur in late July and August and the adults overwinter in their pupal chambers. Very little is known about the duration of larval life; for A. haemorrhoidalis (Fabricius), Roberts (1922, p. 316) suggests 3 or 4 years; for the common American species, there is evidence that indicates a shorter life cycle. The undivided prosternum (prst, fig. 31, c), places Athous near Hemicrepidius, Lepturoides, Elathous, Limonius, and certain species of Ludius. Generic distinction is most readily obtained if the larvae of Athous are considered in two artificial groups as follows: Group I, caudal notch large (figs. 34, 1; 35, c; 36, a) ; eyes present. Group II, caudal notch small (fig. 34, c, d) ; outer prongs of urogomphi much longer than inner prongs; dorsum yellow or yellowish brown. Group I resembles the larvae of Hemucrepidius, but is distinguished by the presence of eyes. Group II differs from Lepturoides larvae in color, and from the larvae of Elathous, Limonius, and allied Ludius by the long outer prongs. Larvae with large, deep pits or prominent trans- verse rugae on the abdominal mediotergites (figs. 34, e; 35, ¢; 36, c; 37, b) are either Athous or Hemicrepidius, depending upon the pres- ence or absence of eyes. The Asiatic Pleonomus has been reported by Ghilarov (1937, p. 635) as inseparable in the larval stage from the European Athous, but specimens were not available for examina- tion in the present study. 14.On the basis of larval characters the European Athous niger (Linnaeus ) and a larva believed to be Athous hirtus (Herbst) are typical Hemicrepidius and are discussed under that genus. NO. II LARVAE OF THE ELATERID BEETLES—GLEN I7I Excluding Athous niger (Linnaeus) and its allies, all the Athous larvae examined had the following characters in common: Pre- sternum of prothorax undivided. Urogomphi bifid, usually with prongs subequal or outer prongs longer than inner prongs. Dorsum of ninth abdominal segment with median sulcus, without setae on central area, and with prominent “teeth” (sometimes sharp) on lat- eral aspects. Nasale tridentate at tip, denticles subequal. Subnasale of many fine teeth. Eyes present. Two unpaired lateroepicranial setae on each gena. Gula of moderate length, but narrow. One “sensory” appendix on second segment of antenna. Mandible usually with well- developed retinaculum but lacking other teeth or toothlike expansions. Definite impressions on mediotergites of mesothorax and meta- thorax. Abdominal mediotergites with long impressions, reaching at least three-quarters of the distance to the mediodorsal suture; mediotergite sometimes with coarse deep pits or prominent trans- verse rugae. KEY TO “SPECIES GROUPS”? AND ISOLATED SPECIES OF ATHOUS ForGandalenotchastiall (hie, SA Gad) icioek ae oe aes comin © foe aerate ee choise 2 Caudal notch large (figs. 34, h; 35, c) or moderately large (fig. 36,a@).. 3 2. Abdominal mediotergites with numerous transverse rugae (fig. 34, e) ; INorthpAweniGaescisss a. scl wac.gnetnn ones cant the rufifrons group (p. 173) Abdominal mediotergites without transverse rugae; Europe............ Bae Ser cay aan cin) ciapexerre ey arayefazs atta, stigpa hi anereite aleyee ysinic the vittatus group (p. 171) 3. Urogomphi with outer prongs much longer than inner prongs (fig. 34, h) ; abdominal mediotergites without coarse punctures or prominent MUpaels JLurOpes cuiatrac solstice Liseies A. mutilatus Rosenhauer (p. 174) Urogomphi with outer prongs not longer than inner prongs (figs. 35, c, d; 36, a) ; abdominal mediotergites with coarse, prominent sculp- (Ana! | GREENS 2o SCH) lamas deolo.o oO MEMOoS one an eoen ae dou sac nosaE 4 4. Caudal notch only slightly narrowed posteriorly (fig. 35, c, f) ; mandible (fig. 35, @) with medial expansion (ex) in region of penicillus; INorthivAmeri Cakieetciacs ache os ioisis tieieyesiereioree the cucullatus group (p. 175) Caudal notch considerably narrowed posteriorly (fig. 36, a, d) ; mandible without medial expansion at penicillus ; Europe and North America. . Peo Na Nasa ECallo ay SELES) SSH OSC gbIR SG avers Seea ER aol gos the undulatus group (p. 177) THE ATHOUS VITTATUS GROUP FIGURE 34, a-c PROVISIONAL KEY TO SPECIES Inner prongs of urogomphi (fig. 34, a, c) with prominent posterior tubercle; outer prongs with small denticle on inner aspect (sometimes absent through erosion) ; fine sharp points on “teeth” (to) along lateral mar- gins or dorsum_or ninth abdomunal seoment.-sq2s secs oe eee eee 172 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Inner prongs (fig. 34, b) strongly convex posteriorly, but without tubercles as described above; without fine sharp points on “teeth” on ninth : fA. vittatus (Fabricius) abdominalasesinentaseen ren ee nein ttre | A. enbfuscus (Mallen) The species placed in this group are enumerated in the key and include A. vittatus (Fabricius), which is regarded by Mequignon (1930, p. 95) as the genotype of Athous. Larvae of all species are reported to occur in decaying wood, but the typical larval habitat appears to be in the soil of grassy areas in and around forests. When such areas have been brought under cul- tivation, the larvae have been reported as damaging domestic plants. Pupation occurs in late July or during August, the adults hibernat- ing in their pupal chambers. This group may be identified by the following combination of characters: Dorsum bright yellow to yellow-brown; caudal notch small; urogomphi (fig. 34, a-c) with outer prongs much longer than inner prongs; abdominal mediotergites finely punctulate, but with- out prominent transverse rugae. Larvae of haemorrhoidalis are reported by Henriksen (1911, p. 275) to attain a length of 24 mm.; the other species only 18 mm. All species have ninth abdominal segment with 4 prominent “teeth” on lateral margins of dorsum, and distance between caudal notch and pleural area approximating one-fifth of total length of segment, ex- clusive of urogomphi. Frontoclypeal area truncate posteriorly. With- out small seta ventrally on basal segment of labial palpus. Up to 5 spinelike setae on each episternum of mesothorax and metathorax. Abdominal mediotergites with impressions reaching to or almost to middorsal suture on second to fifth segments, and with 5 to 8 setae (not definitely paired) in posterior transverse row. Material used in the study of the group.—A. vittatus (Fabricius) : Five larvae from Vemmetofte, Sealand (Denmark), were examined. These are not known to be associated with reared material, but were received fully labeled from Dr. Mathias Thomsen, Royal Veterinary and Agricultural College, Copenhagen. The identification is believed to be reliable. (C.N.C.) A. haemorrhoidalis (Fabricius) : Three larvae, from France, Eng- land, and Denmark, were studied. These were not identified through rearing, but the larva of this species is well known in Europe and the material used is believed to be reliably named. (B.M. and U.S.N.M.) A. subfuscus (Miller): Five larvae were examined. These speci- mens were from Germany (2), Finland (1), and Denmark (2). As NOD Dt LARVAE OF THE ELATERID BEETLES—GLEN 17S far as is known this material was not associated with reared speci- mens, but the species has been reared in Europe and the material available was similar to that figured by Henriksen (1911, figs. 72-74) as A. subfuscus. (C.N.C. and U.S.N.M.) THE ATHOUS RUFIFRONS GROUP Ficure 34, d-f PROVISIONAL KEY TO SPECIES 1. Abdominal mediotergites (fig. 34, e) with coarse, conspicuous, transverse rugae; central and eastern United States and Canada.............. 2 Abdominal mediotergites with shallow, less conspicuous rugae; along the coast and mountains from Alaska to California.................... a ee isis fore [ata c ouwbakelcncies crexcnar eal evecare ai bieie wei A. pallidipennis Mannerheim 2.15 Urogomphi (fig. 34, d) with inner prongs (ifr) somewhat ensiform; usually in decaying wood, rarely in leaf litter..A. rufifrons (Randall) Urogomphi (fig. 34, f) with inner prongs (ifr) more subquadrate; known only from damp decaying forest litter....A. brightwelli (Kirby ) This group is described from the larvae of three North American species. Larvae of A. rufifrons have been collected in New Bruns- wick, Canada, and in Minnesota, Maryland, and Delaware, but larvae of A. brightwelli have been taken only in Pennsylvania although the adults are known as far west as Indiana (Blatchley, 1910, p. 759). Larvae of the western 4. pallidipennis are known only from Walla Walla, Wash. These are all forest species and probably are chiefly predaceous. Rearing notes for rufifrons show that larvae collected in April have matured to adults in May, suggesting that the normal time of pupa- tion is early summer. Larvae of this group are characterized by the yellow-brown color, the small caudal notch, urogomphi (fig. 34, d, f) with outer prongs much longer than inner prongs, and abdominal mediotergites (fig. 34 e) with an abundance of transverse rugae. Mature larvae may exceed 20 mm. in length. Ninth abdominal segment with 4 “teeth” (the most posterior “tooth” usually large and sharp) on each side of dorsum, and distance between caudal notch and pleural area about one-fifth to one-fourth total length of segment, 15 The characters used for separating rufifrons and brightwelli are entirely provisional. Only larval exuviae of brightwelli were available for examination and further study of whole larvae is necessary before better key characters can be given. The data on habitats probably are insufficient to serve as a basis of separation. 174 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT exclusive of urogomphi. Frontoclypeal area truncate posteriorly. Without seta on basal segment of labial palpus. Up to 6 (usually 3 or 4) spinelike setae on each episternum of mesothorax and meta- thorax. Abdominal mediotergites with impressions usually reaching to middorsal line on second to sixth segments, and with 5 to 10 unpaired setae in an irregular posterior transverse row. Material used in the study of the group—Reared material was available for rufifrons and brightwelli. W. J. Brown, of Ottawa, identified the reared adults. A. rufifrons (Randall): Ten examples were examined, including the exuviae of three reared specimens. Eight of the total were from Minnesota, one from Delaware, and one from Maryland. Material associated with available reared adults is listed below. 8; Carter County, Minn.; Apr. 8, 1933; 2 adults emerged May 3, and May 4, 1933, respectively; R. Glen. (C.N.C.) 1; Centerville, Del.; Apr. 12, 1930; adult emerged May 17, 1930; C. A. Thomas. @eaG>) A. brightwelli (Kirby): Only the larval exuviae of two reared specimens were available for study. 2; Cornwells, Pa.; April 1927; €. A. Thomas. (Pai@ and’ CN.) A. pallidipennis Mannerheim: Eight larvae from Walla Walla, Wash., were examined. These were collected and identified by M. C. Lane. (U.S.N.M.) ATHOUS MUTILATUS Rosenhauer FIGURES 12, C; 34, g,h This European species was studied from larvae collected in Ger- many and Denmark. Specimens collected from decaying elms were reported reared to maturity. Larva of A. mutilatus are not closely allied to other known Athous larvae and may be distinguished by the following characters: Caudal notch large, only slightly narrowed posteriorly; urogomphi (fig. 34, g, h) with outer prongs much longer than inner prongs; outer prongs corniform, tips sharp; and abdominal mediotergites punctulate, but lacking transverse rugae and coarse pits. The larvae examined measured up to 13 mm. in length. Dorsum pale yellow to yellowish brown with head and thorax distinctly darker. Ninth abdominal segment with 3 rounded “‘teeth” on each lateral margin of dorsum, and distance between caudal notch and pleural NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 175 area approximating one-sixth of total length of segment, exclusive of urogomphi. Frontoclypeal area bluntly rounded posteriorly. With- out setae ventrally on basal segment of labial palpus. Third segment of antenna as long as second segment. With 3 or 4 spinelike setae on each episternum of mesothorax and metathorax. Abdominal mediotergites with impressions extending to the middorsal suture on second to eighth segments, inclusive, and with 5 to 7 prominent setae in posterior transverse row, none definitely paired. Material used in study—F¥our larvae were examined, one from Dyrehaven, Denmark, and three from Lossnig (near Leipzig), Ger- many. The German specimens were labeled “reared,” but adults and larval exuviae of reared specimens were not available to the writer. (Canadian national, U. S. National Museum, and van Emden collec- tions. ) THE ATHOUS CUCULLATUS GROUP FIGURES 12, a; 35 KEY TO SPECIES 1. Ninth abdominal segment (fig. 35, c) with large deep pits on dorsum; larva usually dark brown; urogomphi with outer prongs slightly shorter than inner prongs (fig. 35, c, d) ; Manitoba and Minnesota tO. Atlantic: COAStamiicis scilete wiovere toler rete ohetstoem ate cieteicnalens @ cucullatus (Say) Ninth abdominal segment (fig. 35, f) with small, sparse punctures on dorsum; larva yellowish brown; urogomphi with prongs subequal HTML CIIS Chea Ae Mc tecra Mteyetole ore wicta- Rie cha opauevene falar stats oisio(Qo 6. asect join vo/e a) sine, spate 2 2. Abdominal mediotergites with the 2 most medial setae occurring close together in a definite “paired” arrangement (fig. 35, e¢) ; known from Illinois and Tennessee to Atlantic coast.............. scapularis (Say) Abdominal mediotergites with the 2 most medial setae well separated, not more than “semipaired”; British Columbia and Alberta to Cali- homme, ahacl AsHLANIE, oo goausaouscd0sb0nbedd0C0bnAE nigropilis Motschulsky On the basis of larval characters, the three North American spe- cies listed above have been placed in the same group, but scapularis and nigropilis are most closely allied. The larvae of all species have been identified through rearing. These are all forest forms, the larvae occurring chiefly in decaying wood. Rearing records and adult captures indicate that pupation normally occurs in early summer, the adults being most numerous in July. Larvae of the cucullatus group differ from all other known elaterid larvae in the mandible (fig. 35, a), which bears a prominent medial expansion in the region of the penicillus. However, the larvae may be more easily recognized by the following combination of characters: Caudal notch large (cn, fig. 35, c, f), only slightly narrowed pos- 170 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III teriorly ; urogomphal prongs subequal, or inner prongs (ipr) longer than outer prongs, outer prongs (opr) with bluntly rounded tips; prominent transverse rugae on abdominal mediotergites (fig. 35, e), especially on third, fourth, and fifth segments. Mature larvae usually measure 16 to 19 mm. in length. Dorsum dark brown (cucullatus) or yellowish brown. Ninth abdominal seg- ment (fig. 35, c, f) with (cucullatus) or without conspicuous, large pits on dorsum, with 3 prominent “teeth” (usually blunt) on each lateral margin of dorsum, and with distance between caudal notch and pleural area approximating one-ninth to one-sixth total length of segment, exclusive of urogomphi. Frontoclypeal area truncate posteriorly. Without setae on ventral aspect of basal segment of labial palpus. Up to 9 (usually 6 or 7) spinelike setae on each episternum of mesothorax and metathorax. Abdominal mediotergites (fig. 35, e) with impressions reaching to middorsal suture on some segments ; transverse rugae on anterior segments give place to circular pits on posterior segments; and usually 5 or 6 prominent setae in posterior transverse row. Material used in the study of the group.—As stated above, reared material was available for all three species. All the reared adults were identified by W. J. Brown, Ottawa. A. cucullatus (Say): The 12 specimens examined were collected in Minnesota (1), Illinois (6), Mississippi (1), and Pennsylvania (4). In four instances larval skins were preserved with reared adults, but only one of the available whole larvae was directly associated with a reared specimen, through being collected at the same time and place. Material for which the reared adults are available is detailed below : 1; St. Paul, Minn.; adult emerged June 4, 1933; R. Glen. (C.N.C.) 1; Hummelstown, Pa.; (no date); J. N. Knull. (U.S.N.M.) 2; Cornwells, Pa.; April 1927; 1 reared; C. A. Thomas. (Pa.C.) 1; Crowell Woods, Pa.; adult emerged June 25, 1930; C. A. Thomas. (Pa.C.) A. scapularis (Say): Nine specimens were examined. These were from Illinois (3), Tennessee (1), Maryland (1), Pennsylvania (2), and Delaware (2). The larval skins of 3 reared specimens were studied, but none of the available whole larvae were directly asso- ciated with the reared material, being from different localities. Ma- terial for which the reared adults are known to be available is listed below : 1; Reelfoot, Tenn.; adult emerged May 21, 1936; A. P. Arnason. (C.N.C.) 2; Centerville, Del.; Apr. 12, 1930; 1 adult emerged June 21, 1930, other speci- men pupated June 10, 1930; C. A. Thomas. (Pa.C.) NO. II LARVAE OF THE ELATERID BEETLES—GLEN LI. A. nigropilis Motschulsky: Eight specimens were examined, in- cluding the exuviae of three larvae reared to adults. All were col- lected from the same cottonwood stump. 8; Lethbridge, Alberta; June 5, 1935; 3 reared adults emerged June 23, July 12, and July 12, respectively; R. Glen. (C.N.C.) THE ATHOUS UNDULATUS GROUP FIGURE 36 KEY TO SPECIES Urogomphi (fig. 36, a@) with outer prongs slightly shorter than inner prongs; dorsum of ninth abdominal segment with conspicuous pits anteriorly ; mesothorax and metathorax with well-developed impressions on medio- LOT OILES a tees a list arcu eePa san ete eeeteneio ake Sener he eer cee ve A. undulatus (DeGeer ) Urogomphi (fig. 36, d) with prongs subequal; dorsum of ninth abdominal segment with conspicuous pits both posteriorly and anteriorly ; meso- thorax and metathorax with very short impressions on mediotergites... 2 356 Sac RGAS RONG Eis RS RAT BP TORS See BN CRE Sa ir A. villosus (Geoffroy) The larvae of Athous (Harminius) undulatus (DeGeer) and Athous villosus (Geoffroy) (=A. rhombeus Olivier) are sufficiently closely allied to be placed in the same group until the larvae of other related species are available for examination. Both species occur in Europe and undulatus is reported by Van Dyke (1932, p. 368) also to inhabit the Hudson Bay and Lake Superior regions of North America. The larvae are found in decaying wood and are believed to be predaceous. Larvae of this group superficially resemble those of Athous cucul- latus (Say), but differ in characters of the mandible, sculpture of abdominal mediotergites (fig. 36, c), and shape of the caudal notch Gist 36,0, d). Mature larvae usually exceed 20 mm. in length. Dorsum is dark brown to brownish black, with large, deep pits, sometimes confluent in second to sixth segments. Venter pale yellow. Caudal notch at least moderately large, subcircular, much narrowed posteriorly. Urogomphi (fig. 36, a, b, d, e) with subequal prongs, or with inner prongs longer than outer prongs; outer prongs (opr) with bluntly rounded tips. Ninth abdominal segment with large, deep pits on dorsum; sometimes (villosus) with 2 fine setae anteriorly on dorsal plate; with 3 promi- nent blunt teeth on each lateral margin of dorsum; and distance be- tween caudal notch and pleural area approximating one-eighth of total length of segment, exclusive of urogomphi. Frontoclypeal area truncate (villosus) or broadly rounded (undulatus) posteriorly. Man- I 78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III dible without medial expansion in region of penicillus (thus differ- ing from A. cucullatus group, fig. 35, a). Without setae on ventral surface of basal segment of labial palpus. Up to 5 spinelike setae on each episternum of mesothorax and metathorax. Abdominal medio- tergites (fig. 36, c) with impressions of variable length, sometimes appearing to extend to middorsal suture on second to fifth segments, but usually difficult to distinguish impressions from confluent ad- joining pits; from 6 to Io setae in posterior transverse row, usually some arranged in pairs. Material used in the study of the group.—A. undulatus (DeGeer) : Four larvae from Finland were examined. The larva of this species has been known in Europe for many years and the writer believes that this material is reliably identified although probably not directly associated with reared specimens. (C.N.C. and U.S.N.M.) A. villosus (Geoffroy): Four specimens from Denmark were ex- amined, including the exuvium of one specimen said to have been reared. The adult of this specimen was not available to the writer, but the label stated ‘“‘adult det. by August West.” It is also known that August West (1937, p. 484) bas reared the larva of Athous villosus from Dyrehaven, Denmark. 1; Bognaes, Denmark; (no date); reared; J. P. Kryger (U.S.N.M.; adult probably in A. West’s collection, Denmark). Other specimens examined are deposited in the Canadian national and the U. S. National Museum collections. Genus HEMICREPIDIUS Germar FIGURES 12, d, f; 37, a-c Species of Hemicrepidius are commonly recorded in the older North American literature under the generic name Asaphes Kirby. Characters of this genus have been based upon the larvae of H. memnonius (Herbst), H. hemipodus (Say) (= decoloratus Say), H. bilobatus (Say), H. carbonatus (LeConte), and a larva, the reared adult of which was identified as “Hemicrepidius sp. near carbonatus (LeConte).” Larvae of the European Athous miger (Linnaeus) and a larva which might be Athous hirtus (Herbst) have been included with Hemicrepidius in the present discussion since these possess typical Hemicrepidius characters. H. hemipodus and H. carbonatus are the only American species that have been collected in numbers suitable for adequate character- ization of their larvae. Other species seem to be rare, with the re- NO. II LARVAE OF THE ELATERID BEETLES—GLEN 179 sult that single specimens are taken and reared to adults leaving only the larval exuviae for examination. Additional material and further study are required for sound selection of key characters and for the definition of “species groups” within the genus. A single larva of Hemicrepidius sp. was collected from a decaying cottonwood stump on the banks of the Old Man River at Lethbridge, Alberta. Decaying wood may be the typical habitat of this species, but the larvae of the other species are known to be primarily soil inhabiting, preferring moist soils such as woodland meadows and only rarely occurring in forest litter or in decaying wood. H. memnonius (Herbst) has been reared from a New Brunswick garden, and from leaf litter in Pennsylvania. However, the species is recorded as far west as Alberta. H. bilobatus (Say) and H. hemipodus (Say) appear to be more truly eastern, not being recorded from west of the Great Lakes. H. hemipodus is the only species commonly encountered in the larval stage. It has been taken from cornfields and other cultivated or pasture land from Maine to Illinois, but apparently it does not cause very severe crop injury. Forbes (1892, p. 39) suggests that this may be due in part to the early date at which larval activity ceases, pupation occurring normally in May or June. The newly formed adults appear to leave their pupal cells immediately. Similar habits of pupation and adult emergence are reported (Henriksen, 1911, p. 274) for Athous niger, and rearing records of other Hemicrepidius larvae indicate that pupation normally occurs relatively early in the summer. This biological character is common to Hemucrepidius, Lep- turoides, several American species of Athous, and possibly Elathous. However, Blatchley (1910, p. 770) reports finding adults of H. mem- nonmus in Indiana from June 8 to December 27, hibernation occurring beneath the bark of red oak logs. On the basis of larval structure, Hemicrepidius is very closely re- lated to Athous. Distinction is most readily obtained through the absence of eyes in Hemucrepidius. Other important characters are: Dorsum yellowish brown, never dark brown; large caudal notch; uro- gomphal prongs (fig. 37, c) subequal, or outer prongs slightly longer ; abdominal mediotergites (fig. 37, b) with conspicuous transverse rugae, and with impressions reaching to middorsal suture in some segments. The urogomphal prongs show considerable individual varia- tion ; in typical specimens each prong has a short, sharp tip, but this is frequently worn away, especially on the outer prongs, giving them the smooth, rounded appearance found in Athous cucullatus (Say) and its allies. 180 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Only large and moderate-sized species are known, the mature larvae usually exceeding 20 mm. in length. Dorsum of ninth abdominal seg- ment (fig. 37, ¢) with median dorsal groove (mg), without setae on the central area, frequently with noticeable pits, and with 3 or 4 prominent “teeth” (to) on each lateral margin. Nasale tridentate at tip. Subnasale of many fine teeth. Two unpaired lateroepicranial setae on each gena. Gula short and narrow. One “sensory” appendix on second segment of antenna. Mandible (fig. 37, a) usually with relatively short retinaculum (ret). Presternum of prothorax undi- vided. Mesothorax and metathorax with distinct impressions on mediotergites, and up to 8 spinelike setae on each episternum. Each abdominal mediotergite with 5 or more prominent, unpaired setae in posterior row. PROVISIONAL KEY TO SPECIES OF HEMICREPIDIUS f. Prom, Europe... cc sae ecto 0.6 aus: 0 «'etore eisle.ciarecdlete wie a see sein eye & Shee ee ee 2 From: NorthvAmenicass 005 «2s cces ceca cts enes tt cect eee 3 Distance between caudal notch and pleural area of ninth abdominal seg- ment about one-seventh total length of segment, exclusive of uro- gomphi; frontoclypeal area almost pointed posteriorly.............. sgie SSR aol aoe Met are a Non AIR UE ar eo) Scene enue Athous hirtus (Herbst) (?) Distance between caudal notch and pleural area about one-fifth length of segment; frontoclypeal area truncate posteriorly.................. sah BPS S PING ects Bhs RARER oe Bye ae Conia No Athous niger (Linnaeus) N 3. Inhabiting rotten wood; collected at Lethbridge, Alberta............... cua MNareh Sad Gee aba eua ome fase coals ote Hemicrepidius sp., near carbonatus (LeConte) Inhabitine soil; rarely in’ forest litters. 2.04.0. 06+. otc ein eee eee 4 4. Basal segment of labial palpus with small seta ventrally; western North America (larvae trom: Walla Walla, Wash®)rresessoseeeee eee Be OAR Ao Se On Ott FOOD Ono H. carbonatus (LeConte) Basal segment of labial palpus without seta; east of Rocky Mountains... H. memnonius (Herbst) Ste GRR CHL y aioe eno amo ulbos Sanur H. hemipodus (Say) H. bilobatus (Say) Material used in study of the genus.—Reared adults of the follow- ing species of Hemicrepidius have been identified by W. J. Brown, of Ottawa: memnonius (Herbst), hemipodus (Say), bilobatus (Say), and “Hemicrepidius sp., near carbonatus (LeConte).” Reared material was also available of Athous niger (Linnaeus). H. memnonius (Herbst): The larval exuviae of 3 reared speci- mens were examined, but no reliably named whole larvae were available. 2; Fredericton, New Brunswick; (no date); both reared; R. P. Gorham. (ENG) 1; Cornwallis, Pa.; April 1927; reared; C. A. Thomas. (Pa.C.) NOG ET LARVAE OF THE ELATERID BEETLES—GLEN 181 H. hemipodus (Say): Eight examples were studied, including the exuviae of four reared specimens. Unfortunately, the reared speci- mens were not collected at the same time and place as the whole larvae that were available. 1; Orono, Maine; Aug. 1, 1929; reared; J. H. Hawkins. (U.S.N.M.) 3; Downingtown, Pa.; Apr. 10, 1930; adults emerged May 13, June 2, June 12, 1930; ©. A: Thomas. (Pa:C.) Hi. bilobatus (Say): Only the exuvium of 1 reared specimen was studied. 1; Hulmeville, Pa.; Apr. 13, 1927; reared; C. A. Thomas. (Pa.C.) H.. carbonatus (LeConte): Eight larvae were examined. These were collected at Walla Walla, Wash., by M. C. Lane and identified by him. (U.S.N.M.) Hemicrepidius sp., near carbonatus (LeConte) : Only the exuvium of one reared specimen was available. 1; Lethbridge, Alberta; June 5, 1935; adult emerged July 12, 1935; R. Glen. (GN:G.) Athous niger (Linnaeus): Seven specimens were studied, includ- ing the exuviae of two that were reared. None of the whole larvae was from the same collection as the reared material, but specimens were examined that were used by K. L. Henriksen (1911) in his description of this species. The material was from Denmark (5), Holland (1), and Germany (1). (U.S.N.M.) 2; Denmark; 1895; both reared; A. Ditlevsen. (R.V.A.C.) Athous hirtus (Herbst) (?): Examination of this species was lim- ited to a single specimen from France labeled as “Athous hirtus Hbst. or niger L.?” This larva differed slightly from available speci- mens of A. niger. (van Emden.) Genus CREPIDOMENUS Erichson CREPIDOMENUS QUEENSLANDICUS Blair FIGURES I0, f; 37, d-f This Australian species is the only Crepidomenus known in the larval stage. Specimens were received for study through the courtesy of W. A. McDougall, Assistant Entomologist, Central Sugar Experi- 182 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III ment Station, Mackay, Queensland, and from the British Museum of Natural History, London. According to McDougall (1934, pp. 60, 65, 67; “B sp.’”’) the larvae inhabit the soil of cultivated fields and grasslands; feeding normally occurs only during short periods immediately after each ecdysis; the complete larval life is probably less than 1 calendar year ; and pupation occurs in September or early October, the adults being found in great- est numbers as early as the middle of October. In structure, the larva of C. queenslandicus resembles the larvae of the genus Cryptohypnus, particularly in possessing dorsal postero- epicranial setae (as in fig. 10, g, ped), and medial anterotergal setae (atm, figs. 10, g; 37, d) in thorax and abdomen. It is distinguished by the following characters: Dorsum of ninth abdominal segment (fig. 37, e, f) with sharp “teeth” (to) on the lateral margins, and only 2 setae on the central area; abdominal mediotergites (fig. 37, d) with long impressions and lacking seta number 7 as numbered in Cryptohypnus (fig. 38, f). Larvae examined measured up to 19 mm. in length; dorsum yel- lowish brown to pale chestnut brown; rather densely set with small punctures. Caudal notch large, not narrowed posteriorly. Urogomphi (ur, fig. 37, e, f) usually with prongs subequal in length, inner prongs (ipr) sometimes longer and always slightly more robust than outer prongs (opr). Ninth abdominal segment (fig. 37, e, f) without median dorsal groove, and distance between pleural area and caudal notch approximating one-fifth to one-fourth length of segment, ex- clusive of urogomphi. Nasale with tridentate tip, lateral denticles only slightly smaller than median denticle. Frontoclypeal area rounded posteriorly. Eyes present; surrounded by 4 setae. Three unpaired lateroepicranial setae on each gena. Gula moderately long, but very narrow. One “sensory” appendix on second segment of antenna. Mandible with prominent retinaculum. Presternum of prothorax divided, the median posterior piece being small. Mesothorax and meta- thorax with distinct impressions on mediotergites, and without spine- like setae on episterna. Abdominal mediotergites (fig. 37, d) with transverse branches of impressions reaching four-fifths to five-sixths of distance from longitudinal branches to middorsal suture. Abdomi- nal pleurites large, subequal from first to eighth segments. Material used in study.—Thirty larvae were examined. All were collected at Mackay, Queensland, by W. A. McDougall, who suc- ceeded in rearing this species. (C.N.C. and B.M.) NO. II LARVAE OF THE ELATERID BEETLES—GLEN 183 Genus CRYPTOHYPNUS Eschscholtz FIGURES I0, g; 38 The name Cryptohypnus has been used in the present study for species which, on the basis of larval characters, obviously are allied to Cryptohypnus riparius (Fabricius) (=Hypnoidus riparius), and for which American authors recently have shown a preference for the generic name Hypolithus Eschscholtz. The name Hypnoidus Stephens is restricted to species whose larval characters stamp them as being congeneric with H. dubius (Horn) and generically distinct from riparius and its close allies. Characters given for this genus are based upon a study of the larvae of the five species listed in the key. All are believed to be soil inhabiting, and some are important crop pests. These larvae superficially resemble Melanactes densus and the Ludius nitidulus group but are much more closely allied to the genus Crepidomenus. Cryptohypnus larvae are distinguished by a combina- tion of characters involving setal arrangement, sculpture, and features of the ninth abdominal segment, the following being of primary sig- nificance: Head bearing dorsal posteroepicranial setae (ped, figs. 10, g; 38, a) ; each thoracic segment and first 8 abdominal segments bearing medial anterotergal setae (atm, figs. 10, g; 38, f) ; abdominal mediotergites with impressions and setal pattern as in figure 38, f, setae numbered 1 to g being present in all species; dorsum of ninth abdominal segment (fig. 38, 7) bearing 4 setae on central plate and blunt “teeth” (to) on the margins. The species studied rarely exceed 16 mm. in length. Dorsum yel- low to yellowish brown; finely punctulate. Caudal notch (cm) large. Urogomphi bifid, prongs subequal in length (fig. 38, e, g) or inner prongs longer (fig. 38, 4,7). Ninth abdominal segment without median dorsal groove, and distance between pleural area and caudal notch varies from one-sixth to one-fourth length of segment, exclusive of urogomphi. Nasale (n, fig. 38, b-d) typically tridentate; lateral denticles may be serrate (fig. 38, d) or absent through erosion. Eyes present. Gula moderately narrow. One “sensory” appendix on second segment of antenna. Mandibles of same type as in Ludius aeripennis. Presternum of prothorax of more than 1 piece, the median posterior sclerite being very narrow and often indistinct. Mesothorax and metathorax with distinct impressions on mediotergites; and with I to 3 setae on each episternum, setae usually fine or with only 1 being spinelike. Abdominal pleurites well developed, decreasing in size from first to eighth segments. 184 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III KEY TO SPECIES OF CRYPTOHYPNUS 1. Urogomphal prongs subequal (fig. 38, e, g) ; eastern North America.... BE ME ets eae EOE Laois an OIA aCe abbreviatus (Say) (p. 185) Inner prongs of urogomphi longer than outer prongs (fig. 38, h,7).... 2 2. Nasale (n, fig. 38, d) with several small denticles on each side of base of median tooth; western North America..... funebris Candeze (p. 185) Nasale (n, fig. 38, b) tridentate, lateral denticles sometimes absent through) erosion sacs sent Cire seyret (the riparius group, p. 184) 3 4... OROMN SEMIOPE spaepeisvsicy lev oe wists aye eisk-tol s evelaie oehe le, Were errant riparius (Fabricius) From: ‘North’ America’s 0 5 2o2.8ccre ones ecrars ot thors ec tensusla eherere a aCe ner etarenenS 4 4. Basal segment of labial palpus with 1 seta ventrally......... sanborni Horn Basal segment of labial palpus without setae...... nocturnus (Eschscholtz) THE CRYPTOHYPNUS RIPARIUS GROUP Ficures 10, g; 38, b, h,1 Included in this group are riparius (Fabricius), nocturnus (Esch- scholtz), and sanborni Horn. All live in the soil. C. nocturnus is reported by King (1928, pp. 703-704) as an important pest of grain crops in Saskatchewan, and Evans (1921) suggests that riparius might injure field crops in Scotland. This is a very homogeneous group, well characterized by the urogomphi and caudal notch (fig. 38, , 7) and by the nasale and sub- nasale (fig. 38, b). These larvae seldom exceed 13 mm. in length. The inner prongs of the urogomphi are about twice as long as the outer prongs; the tips of the outer prongs turn slightly inward. The large, U-shaped caudal notch is not narrowed posteriorly except in a few atypical specimens. In the tridentate nasale, the lateral denticles are often weakly developed and sometimes are eroded away, giving a unidentate appearance. The frontoclypeal area is broadly rounded posteriorly. Material used in the study of the group.—cC. riparius (Fabricius) : Eight larvae were examined. These were from Denmark (5), Fin- land (1), Russia (1), and Ireland (1). The specimens studied are believed to be reliably named, because the larva of this species is well known in Europe and some of the available material was used by K. L. Henriksen (1911) in his description of this species. (B.M. and U.S.N.M.) C. nocturnus (Eschscholtz): Thirteen examples were studied, in- cluding the exuviae of four reared specimens. The reared adults were identified by W. J. Brown, of Ottawa. 10; Barnwell, Alberta; June 6, 1935; 2 adults emerged Aug. 8 and Aug. 9, 1935; R. Glen. (C.N.C.) NO. II LARVAE OF THE ELATERID BEETLES—GLEN 185 1; Taber, Alberta; June 6, 1935; adult emerged Jan. 31, 1936; R. Glen and G. F. Manson. (C.N.C.) 2; Turtleford, Saskatchewan; June 10, 1937; 1 adult emerged Aug. 3, 1937; J. V. Brooks. (C.N.C.) C. sanborni Horn: Twenty specimens were studied, all from Churchill, Manitoba. A reared adult was identified by W. J. Brown, of Ottawa. 20; Churchill, Manitoba; June and July 1937; 1 adult emerged in July 1037; W. J. Brown. (C.N.C.) CRYPTOHYPNUS FUNEBRIS Candeze Ficure 38, d, f This western species is distinguished by the numerous denticles on nasale and subnasale (fig. 38, d), and by the frontoclypeal area, which is bluntly pointed posteriorly. The caudal notch is U-shaped and is not narrowed posteriorly. The urogomphi have the inner prongs longer than the outer prongs but less than twice as long. The setal pattern on abdominal mediotergites is given in figure 38, f. Larvae of this species are larger than other known Cryptohypnus, those at hand measuring about 16 mm. in length. Material used in study.—Only 2 larvae were available for ex- amination. These were collected, along with adults of this species, at Walla Walla, Wash., by M. C. Lane, and identified by him. It is not known that larvae of this type have actually been reared to adults, but Mr. Lane has expressed his confidence in the identification made. (U.S.N.M.) CRYPTOHYPNUS ABBREVIATUS (Say) FIGURE 38, a, C, €, 9 This species is widely distributed in the eastern United States and from Newfoundland as far west as Saskatchewan. It is an important pest in fields and gardens. The larva differs from other known Cryptohypnus in having the urogomphal prongs (fig. 38, e, g) sub- equal ; the caudal notch subcircular, or transversely subovate, and con- siderably narrowed posteriorly ; the subnasale (fig. 38, c) with about 7 or 8 denticles ; and the frontoclypeal area (fig. 38, a) rounded poste- riorly. The lateral denticles of nasale are sometimes eroded, giving the appearance of a unidentate structure. Fully distended mature larvae measure up to 12 mm. in length. 186 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Material used in study—Examination was made of 12 examples, including the exuviae of 4 reared specimens. The reared adults were identified by W. J. Brown, of Ottawa. 3; Rhein, Saskatchewan; June 11, 1930; 2 adults emerged July 21 and Aug. 6, 1930; E. Mengering. (C.N.C.) 9; Spencerville, Grenville County, Ontario; June 23, 1938; 2 adults emerged Aug. 26 and Sept. 2, 1938; G. H. Hammond. (C.N.C.) Genus HYPNOIDUS Stephens Ficures 9, f; 10, b; 39 The name Hypnoidus is used here in the restricted sense and does not include the species which are commonly referred to the genus Cryptohypnus. Knowledge of this genus is based upon an examination of larvae of dubius (Horn), musculus (Eschscholtz), and the European mari- timus Curtis. These species are very similar and it might be that they represent only one group within the genus. All are believed to be soil inhabiting and dubius is recorded by King (1928, p. 705) as being of economic importance in Saskatchewan. Kincaid (1900, pp. 374-375) described the larva of musculus from Alaska. The known Hypnoidus larvae rarely exceed 11 mm. in length, are very pale yellow in color and are readily identified by the absence of spinelike setae on the coxae (fig. 39, e), by the mandibles (fig. 39, c,g) and by the pointed, undivided urogomphi (fig. 39, f, h, 1). Some- times, especially in musculus, a slight tubercle is present on the latero- posterior aspect of each urogomphus, which might be interpreted as an extremely small outer prong. The caudal notch (cv) is large, usually transversely ovate and more or less narrowed posteriorly. The ninth abdominal segment lacks the transverse impression, the median sulcus, and the “teeth” on lateral margins, but bears 4 setae on central dorsal area. Nasale (1, fig. 39, a) consists of 3 short, subequal teeth, lateral denticles some- what serrate in maritimus. Frontoclypeal area pointed posteriorly. Eyes lacking. Gula (fig. 39, b) very long and narrow. One “sensory” appendix on second segment of antenna. Galea reduced to a single segment. Mandible (fig. 39, c, g) with small retinaculum (ret), and finely serrate posterior to base of retinaculum. No distinct impressions on mediotergites of mesothorax and metathorax. Without spinelike setae on thoracic episterna. Abdominal mediotergites slightly rugose, NOS Li LARVAE OF THE ELATERID BEETLES—GLEN 187 not punctured, impressions indistinct, and prominent setae unpaired. Abdominal pleurites (i, fig. 39, d) large, sterna (st) undivided. PROVISIONAL KEY TO SPECIES OF HYPNOIDUS Heme CAT Ottat NOTE CAMO Cats re ayansiesrsis, oie sioisvs o, oC ceame tehaeso oct ater e tater sie ereieea eens 2 EE Oiie SU Opey even cere eae toss Siac ies a duel eet e aoe maritimus Curtis 2. Mandible (fig. 30, g) usually with about 5 or 6 fine denticles at base of ELINA CULSITI a: eee peyote sieeve nets arsed asst ae Caste ONS musculus (Eschscholtz) Mandible (fig. 30, c) usually with about 2 fine denticles at base of GE INA CULE mee Cree ert cae One Oto eie a Evins atone dubius (Horn) Material used in study of the genus—Reared adults of H. dubius (Horn) and H. musculus (Eschscholtz) were identified by W. J. Brown, of Ottawa. H. dubius (Horn): Six specimens from Saskatoon, Saskatchewan, were examined, including the exuvium of one reared specimen. (EN:C). 1; Saskatoon, Saskatchewan; May 27, 1924; adult emerged June 25, 1924. K. M. King. H. musculus (Eschscholtz): Four larvae were studied. These were collected by T. Kincaid from Popoff Island, Alaska, and were the specimens upon which he based his description. An adult which was reared from a larva apparently collected at the same time and place is in the U.S. National Museum, but the associated larval skin could not be found. HT. maritimus Curtis: Six larvae from Austria were examined. These were identified through field association with adults of this species. However, the larvae show typical Hypnoidus characters and there is good reason to believe that the identification probably is correct. (Canadian national and British Museum collections. ) Genus EANUS LeConte FIGURE 40, a-e Species of Eanus are commonly recorded in the European literature under the generic name Paranomus Kiesenwetter. Larvae of this group inhabit the litter of the northern and alpine forests and are probably chiefly predaceous. E. decoratus has been reared from Quebec and larvae have been collected as far west as Saskatchewan. According to Brown (1930, p. 163) adults of this species are known from the Gulf of St. Lawrence to Alaska. A larva 13 188 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III believed to be E. costalis was described from Finland by Prof. U. Saalas (1937, pp. 68-72, “Probably Orithales serraticornis Payk.” ; 1938, pp. 53-55), who kindly loaned the writer material for exami- nation. These two species are very similar in structure and are most readily separated on the basis of geographic distribution. However, Brown (1930, p. 162) has indicated that they represent only one of the two distinct species groups included in the genus. On the basis of the larvae examined, Eanus is an isolated genus, readily distinguished by the following characters: Dorsum of ninth abdominal segment (fig. 40, d, e) strongly convex, and without raised lateral margins ; urogomphus (ur) undivided, blunt, and with a strongly sclerotized bar on inner aspect; abdominal pleurites absent or practically so (fig. 40, b, c) ; abdominal mediotergites (mtg, fig. 40, b) with very faint, short impressions, and with only 3 prominent unpaired setae in the posterior transverse row. Eanus larvae are pale yellow; up to 18 mm. in length; with in- conspicuous sculpture. Ninth abdominal segment with small caudal notch, and distance between pleural area and caudal notch about one- fifth length of segment, exclusive of urogomphi. Nasale of 1 sharp tooth. Eyes present; bordered by 2 unpaired setae. Two unpaired lateroepicranial setae on each gena. Gula extremely narrow, area practically closed posteriorly. One “sensory” appendix on second segment of antenna. Mandible (fig. 40, a) very robust with toothlike expansion of dorsal margin of inner face of distal half. Presternum of prothorax divided, the posterior median piece being very small. Mesothorax and metathorax with faint but definite impressions on mediotergites, and without spinelike setae on episterna. Material used in study of the genus.—E. decoratus (Mannerheim) : Eleven examples were examined, including the exuviae of three reared specimens. All were from the province of Quebec and the reared adults were identified by W. J. Brown, of Ottawa. (C.N.C.) Material identified through rearing is listed below: 1; Mount Lyell, Gaspé County, Quebec, 1,500 ft.; (no date); adult found emerged on Oct. 1, 1934; C. C. Smith. (C.N.C.) 2; Cascapedia River, Quebec; adults emerged Sept. 7 and Sept. 20, 1935; M. L. Prebble. (C.N.C.) E, costalis (Paykull) (?): One larva from Suomussalmi, Finland, was all the material available for examination. This was an unidenti- fied specimen that is believed to be E, costalis because of its morpho- NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 189 logical similarity to E. decoratus larvae and because E. costalis inhabits Finland. On the basis of adult characters, Brown (1930) reported these two species to be very closely related. The larva examined is the property of Prof. U. Saalas, Helsinki, Finland. Genus MELANACTES LeConte MELANACTES DENSUS LeConte FicuRE 40, f Melanactes densus is the only species of this genus that is known in the larval stage. It is a western species, but the natural habitat of the larva is unknown, the only specimen available for examination being reared from eggs laid in captivity. On the basis of larval characters this species is a typical member of the tribe Lepturoidini and should not be separated into a distinct tribe as arranged by Leng (1920). Superficial resemblances are found in the larvae of Hemicrepidius, Crepidomenus, and Cryptohypnus, but no closely related genera are known. The larva of M. densus (fig. 40, f) is readily distinguished by its large size, absence of eyes, setal and sculptural patterns, shape of the caudal notch, and the uro- gomphi in which the inner prongs are small and the outer prongs very long and corniform. The larva hitherto regarded by some writers as Melanactes piceus (DeGeer), originally figured by Riley and repro- duced by Blatchley (1910, p. 771, fig. 292), unquestionably is not a Melanactes larva or the larva of any other elaterid. The larva examined was not fully distended, but measured 38 mm. in length and 4.75 mm. in breadth at 5 years of age. Fully distended mature larvae undoubtedly would exceed 40 mm. Dorsum golden brown ; venter paler ; lateral membranes large, creamy white. Caudal notch large, transverse. Urogomphi short, robust, bifid ; prongs corni- form, with sharp upturned tips; outer prongs 3 to 4 times as long as inner prongs. Ninth abdominal segment without a mediodorsal groove, but with a shallow subcircular depression bearing 2 faint, short, para- median impressions; 4 widely separated setae on central dorsal area, posterior setae being longer ; 3 blunt “teeth” on each lateral margin of dorsal plate; tergite continues uninterruptedly on posterior ventral surface for a distance approximating one-eighteenth total length of segment exclusive of urogomphi; sternum large, consisting of 2 sclerites separated anteriorly by a narrow median suture. Nasale of 1 tridentate tooth, median denticle slightly larger than lateral denticles. Frontoclypeal region broadly rounded posteriorly. One small dorsal posteroepicranial seta just laterad to each side of posterior part of IgO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIt frons. Eyes absent. With 4 setae surrounding each eye region. Two unpaired lateroepicranial setae on each gena. Gula short and nar- row. One “sensory” appendix on second segment of antenna. Mandi- ble with well-developed retinaculum. Presternum of prothorax un- divided, but may be deeply incised laterally near apex, almost separat- ing off a small, narrow, posterior median sclerite. Mesothorax and metathorax with prominent long impressions, up to 7 spinelike setae on each episternum. Abdominal mediotergites with impressions reach- ing to middorsal suture; first to seventh segments with conspicuous transverse rugae, eighth segment with pits more nearly circular ; 1 an- terior mediotergal seta present on each mediotergite ; conspicuous setae unpaired on anterior part of segments, but approaching a semipaired arrangement in posterior transverse row. Abdominal pleurites large, decreasing in size from first to eighth segments. Spiracles noticeably widened anteriorly, subpyriform ; situated in anterior part of segments. Material used in study—The only specimen available for exami- nation was a 5-year-old larva received through the courtesy of Dr. W. H. Anderson, U. S. Bureau of Entomology and Plant Quaran- tine, Washington, D. C. 1; Huntington Beach, Calif.; reared from eggs that hatched June 3, 1935; M. W. Stone. (U.S.N.M.) SUMMARY OF TAXONOMIC RELATIONSHIPS In the Lepturoidini, relationships based upon larval characters agree closely with those founded upon careful study of the adults. Recognized subspecies are morphologically indistinguishable in the larval stage; closely allied species appear to be more readily distin- guished as adults than as larvae, as evidenced by the species of Ludius included in the cupreus group and the aeripennis group; but group relationships are revealed as clearly by the larvae as by the adults. The evidence from larval studies should be considered in taxonomic revisions. However, such evidence must be used with exceeding care, because it is based upon a very small proportion of the existing species. Rightly or wrongly, at the beginning of this study the type species were looked upon as the ultimate criteria for the determination of generic status. It was believed that any given larva could be classified to genus by comparing it with genotype material. In practice this proved futile, especially for large genera, because only a very small proportion of the available species of each large “genus” bore evi- dent close relationship to the genotype. The supraspecific unit that stood out clearly as a measure of relationship was the “‘species group,” NO. II LARVAE OF THE ELATERID BEETLES—GLEN IOI DIAGRAM OF GENERAL RELATIONSHIPS BETWEEN LEPTUROIDINE GENERA AND “SPECIES GROUPS” OF LUDIUS Species above the broken transverse line have the nasale single-pointed ; species below this line have the nasale triple-pointed, except in Limonius pilosus (Leske). Genera in parentheses have the urogomphi simple; all others have the urogomphi bifid. The spacing suggests the degree of relationship, and possible annectant groups are indicated by connecting lines. CAUDAL NOTCH LARGE CAUDAL NOTCH SMALL II I EAN rotundicollis ( oe), > bipustulatus Als |S als S| a oe appressus Ss 2 propola 2 |= inflatus S aeripennis triundulatus cupreus (ea) we een we ne ne ee ee ee 5 ee en =e | Pe ee we ee oo nw ew ee ew on ew oe ee eee is semivittatus fallax S edwardsi Ay nitidulus es sjaelandicus = CRMVPTO= CRE PI- = HYPNUS DOMENUS Ss a s ive) 100 = ane rrh B z pyrrhos S w MELANACTES limoniiformis ra) a resplendens ZS ~ HEMICREPIDIUS---ATHOUS || ELATHOUS---LIMONIUS = (cucullatus) = (undulatus) || ATHOUS--LEPTUROIDES oe (mutilatus) (vittatus ) Fal (rufifrons ) 2 divaricatus a rary (HYPNOIDUS) 192 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. GET of which several to many were apparent within each of the larger genera as recognized in the current taxonomic literature. In effect, each genotype was representative of but one “species group.” Some of the larger genera are too complex to be adequately char- acterized. This is particularly true of Ludius and Athous and to a lesser degree of Limonius. On the basis of the material studied, the following genera appear to be sufficiently homogeneous to per- mit of characterization: Crepidomenus, Cryptohypnus, Hypnoidus, Eanus, Elathous, Lepturoides, Hemicrepidius, and Melanactes. How- ever, larvae of only one to five species were available for each of these genera and insufficient representation may account for this apparent homogeneity. In general, the taxonomic value of any character or set of char- acters changes from one “species group” to another. However, throughout the tribe studied certain characters have maintained a high ordinal value. Simple urogomphi and bifid urogomphi have not been found within the same genus; and the type of caudal notch, the general type of prosternum, and the general type of nasale have varied comparatively little between species of the same “species group.” Accordingly, these are regarded as primary group char- acters, their relative importance being as in the order stated. On the basis of these characters the general relationships between the “spe- cies groups” of Ludius and the other lepturoidine genera are charted on page IQI. THE GENUS LUDIUS The “species groups” of Ludius that have been erected on the basis of larval characters agree closely with those independently established by Brown (1935, 1936, 1939), who studied the adults of the American species. Where adequate larval material was avail- able, Brown’s groups were invariably substantiated, but where the larvae of only a very few species were known or where the material was not suitable for a complete examination, the larval study sug- gested fewer groups than Brown had erected. The writer agrees completely with the following statement (Brown, 1935a, p. 1): “The genus Ludius, as now recognized in the American literature, is not a true genus in any natural sense, but is, rather, a heterogeneous group of species the natural affinities of which are not known.” Similar conclusions have been reported by Blatchley (1910, p. 763) and by Van Dyke (1932, pp. 389-390), who studied the adults, and by Hen- riksen (1911, p. 258), who described the larvae of seven species. NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 193 Brown has done much toward establishing the natural affinities of many of the species now included in the genus. However, no proposals have yet been made for reducing the genus to a more homo- geneous assemblage. The larval record is much too incomplete to form a basis for sound revision. In fact, since only 10 percent of the species are known in the larval stage, it is dangerous even to attempt an appraisal of specific relationships. However, a brief appraisal of the larval evidence is presented with the hope that it might encourage further efforts in this direction. The 44 species of Ludius studied have been arranged into 17 dis- tinct “species groups,’ some groups being represented by a single species. For purposes of identification, these groups are best con- sidered in the four major sections (I, II, III, 1V) shown on page 191. However, the natural affinities ot Ludius sjaelandicus (Muller) are believed to lie nearer to several species in section III than to the other species in section I. Within section II some of the “species groups” are very closely related, e.g., inflatus and aeripennis, semivit- tatus and edwardsi; whereas propola, triundulatus, and fallax are not quite so closely allied; and bipustulatus, rotundicollis, and nitidulus are relatively isolated. In contrast, the Ludius species that fall in section III are not more closely related to one another than are the recognized genera Elathous, Limonius, Athous (pars), and Leptu- roides, which also fall in that section. The larval characters suggest that for taxonomic purposes the genus Ludius might be revised by making the following subdivisions (genera ?): SUBDIVISION I: the cupreus and appressus groups; the relationships of appressus (Randall) seem to lie with cupreus. SuBpIvision 2: all Ludius included in section II; the nitidulus group might form the nucleus of a distinct subdivision connecting with the genera Cryptohypnus and Crepidomenus. SuBDIVISION 3: L. divaricatus (LeConte) and its allies; this is an isolated species without close affinities to any known group in the Lepturoidini. Suppivision 4: L. sjaelandicus (Miller) and allied species, if any exist. SUBDIVISION 5: the pyrrhos group. SuBDIVISION 6: the limoniiformis group. Suppivision 7: L. resplendens (Eschscholtz) and its allies; the possibility of removing resplendens to the genus Elathous warrants consideration. Subdivisions 4 and 5 have a somewhat common bond in the unusual condition of the spiracles in the eighth abdominal segment, and in setal and sculptural characters. Likewise, subdivisions 4, 5, 6, and 7 all have the urogomphal prongs subequal and certain other characters in 194 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIt common. However, the union of two or more of these groups pro- duces an assemblage that cannot be adequately characterized to per- mit ready separation from the other genera included in section III of the diagram on page 191. This attempt to reduce the genus Ludius to definable limits is merely suggestive. The larval record is too incomplete to provide the necessary perspective, but it might supplement the evidence from adult studies. There are two alternatives to the suggested subdivi- sions given above: make every “‘species group” a new genus, or leave the genus in its present indefinable state. From the purely phylogenetic viewpoint, it is doubtful if genera are completely de- finable. Therefore group limits for taxonomic purposes are very difficult to draw. Species seem to be well defined in those parts of the genus that Brown has revised. His species have been substantiated wherever adequate larval material was available for study. However, very detailed examination frequently was necessary to discover the sepa- rating characters of closely allied forms. Brown’s species concept appears to be valid, but any narrower concept could not be supported by larval evidence. As more material becomes available, it is quite possible that larval characters will not be found to separate all species that Brown has recognized. Such a result can be expected, since distinctions that are evident in the definitive adults might be ob- scured in the developmental stages. Ludius rotundicollis (Say), as currently recognized, appears to in- clude more than one species. The larvae from the eastern States and those from the western States are morphologically distinct and, therefore, should be considered as distinct species. Tue AtHous CoMPLEX In contrast to the genus Ludius—which has been allowed to de- velop into an indefinable assemblage—other parts of the Lepturoidini have been separated into smaller, closely related genera. Thus Elathous, Hemicrepidius, and Lepturoides are very closely allied to the genus Athous as recognized at present, but Athous itself is poly- morphic. As illustrated on page 191, sections III and IV, Athous consists of two main parts which are less closely related to each other than to other established genera. The failure of European workers to recognize the genus Hemicrepidius in their fauna is due to its*similarity to Athous. The European Athous niger (Linnaeus) and its allies are typically Hemicrepidius. Whether Hemicrepidius NO. II LARVAE OF THE ELATERID BEETLES—GLEN 195 should be recognized as a genus or merely as one or more “‘species groups’ under Athous is open to question. On the basis of larval characters Hemicrepidius is not more distinct from Athous than are the “species groups” of Athous (enumerated in section IV) from one another. The other half of Athous, as represented by the vittatus and rufifrons groups, is closely allied to Elathous and Lepturoides. In fact, Lepturoides is not more distinct from these groups of Athous than these are from each other. Elathous, on the other hand, appears to fill the gap between Limonius and Athous (pars) and probably should be retained as a distinct genus. The rather fragmentary larval record suggests, therefore, that Athous be revised into two genera, as in sections III and IV and that Hemucrepidius might be included with the one part and Lepturoides with the other, or these retained as distinct genera as at present. Elathous should remain as a genus. The writer is fully in accord with Van Dyke (1932, p. 356), who states that Elathous bicolor (LeConte) should not be included with the genus Leptoschema Horn. On the basis of larval characters Leptoschema belongs in the subfamily Elaterinae. THE GENUS LIMONIUS Limonius is less complex than Ludius and Athous. This is shown on page IgI, the genus being confined to section III. However, the larvae exhibit considerable polymorphism, mainly in characters of the mandibles, nasale, and urogomphi. At present, it would appear premature to erect genera on these characters. However, if any re- vision of the genus were to be made on the basis of larval characters the following subdivisions are suggested : SUBDIVISION 1: the aeneoniger group; these larvae are well characterized by the mandibles, nasale, subnasale, and small outer urogomphal prongs. Suppivision 2: L. pilosus (Leske) and its allies; the single-pointed nasale, the subnasale, and the mandibles isolate this larval type. SUBDIVISION 3: includes all remaining species of Limonius known in the larval stage. This subdivision brings together rather diverse forms, but the chief difference is that pectoralis, aeger, and confusus (?) all have very small outer prongs whereas the members of the canus group have well-developed outer prongs. However, this character does not appear to have high ordinal value, especially since aeger is obviously related to the canus group through several other characters. THE GENERA HYPNOIDUS AND CRYPTOHYPNUS There has been a tendency in recent years for taxonomists to suppress the genus Cryptohypnus and to place its species under 196 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Hypnoidus. According to Hyslop (1921), Elater riparius Fabricius was designated by Westwood in 1840 as the type of Hypnoidus Stephens; in 1859 the same species was designated by Thompson as the type of Cryptohypnus. If riparius be used as the type of Hypnoidus, then species such as H. dubius (Horn), H. musculus (Eschscholtz) and H. maritimus Curtis must be placed in a distinct genus. On the basis of larval characters, riparius and its allies defi- nitely are not congeneric with H. dubius and its allies.° The latter group have simple urogomphi, a I-segmented galea, a very character- istic mandible, the prosternum undivided, and several other char- acters which separate them widely from the species which commonly have been placed in Cryptohypunus. THE TRIBE LEPTUROIDINI Without a careful study of the larvae of other tribes of the Pyro- phorinae, the writer cannot appraise properly the status of the Lep- turoidini. However, it would appear to be a reasonably well-character- ized group. Crytohypnus and Melanactes, which Leng (1920) placed in the tribes Hypnoidini and Melanactini, respectively, properly be- long in the Lepturoidini. The most aberrant genera are Eanus and Hypnoidus. These are extreme forms and appear to be quite isolated within the tribe. It may be that larvae eventually will be found to link these genera more closely to the other Lepturoidini. List OF SPECIES STUDIED This list is designed to serve two main purposes: First, as an index to the principal descriptive sections on each species ; and second, as a ready reference to the continental representation and the general reliability of the material used. The genera are listed alphabetically and the species alphabetically under each genus. The following symbols are used: Aus. for the continent of Australia; Eur. for Eurasia; N.A. for North America; * for species of which reared adults and their larval exuviae were available to the writer; # for species whose identification was known to be or believed to be based upon rearing, but reared adults could not be obtained for confirma- tion; x for species whose identification was believed to be fairly reliable, but not based upon rearing as far as was known; and o for species whose identification was very uncertain. Symbols in paren- 16 Support of this conclusion is given in a recent critical review based upon adult characters. (Vide M. C. Lane, “Some Generic Corrections in the Ela- teridae, II,” Proc. Ent. Soc. Washington, vol. 50, No. 8, pp. 221-223, November 1948. ) NO. II LARVAE OF THE ELATERID BEETLES—GLEN 197 theses indicate that larval material was not available to the writer and all knowledge of such species was obtained from the literature. +4 #0 * KM HHO HHOH #HO * HK & HH HH eH H HOM HH eH ® HHH HHH * HH He HH MH Oe OX Species Distribution Page oe brightuellvm GUEDY)) wets ee cies eniee INiAew i eee ee 173 euerioums (SESA) ceéccoeseoaseeccccobcc INA id eerie caren 175 4 haemorrhoidalis (Fabricius) ........... 1 | ee ed ee 171 SME IESE ELL CLDSU) en tarseky ocak Soee lock OT hater Cae eee ht. 180 “ mutilatus Rosenhauer ................-. 1 Dy bbe Aer e e ee eAe 174 MERIC Cr CMANNACUS )he-cegwisies « cieitbern aay steee ire soe eee 180 <)> moropilis Motschulsky, <...2 2.02 cs- eee: ING Acts eeisineicnn cee 175 “ pallidipennis Mannerheim ............. INE Aer Be ye) een 173 o agiaaes (Geib) Sesoocsocdooboe onc INGAG Sins ae ee cies 173 SU SCOPILONUS CSAY))) sys sicia cvstestieis. 0 avers © 0isitvone INRA. Saket atek 175 ee SUD TUS cusE (MINERS aiicclec oteth oes eee Bur. ees eee 172 epamorng ((DaEES2))> Gososonscoacuodeos IDI, we INAS a5o060c 177 Go hileGeS (CEST). Soocoouapesoedoe ac ura Beene 177 So eas (CRESS) pocebooucuonoboone eee eyes eee 172 Crepidomenus queenslandicus Blair ............ AISs Ware eee 181 CTE Ems GUOTEVIGINS, (SAY). assesses o/s sos INDAY: psec oe 185 SaPnona WOE ohoc0bg0e0s5s00006 NAG nae eee 184 is junebris CandeZe joc5.04 86. 2 562 2 NWAG anh pe eees 185 a nocturnus (Eschscholtz) ........ INGA os ceva isis. ot rete 184 ss riparius (Fabricius) ...........- UB Bb eRe: gene ee 184 Hanusecostalisn Geay kttll)e meee: ecisioeseec: Eur ees anon eee 188 ““ decoratus (Mannerheim) ............... INGACHRR aban sees 187 ENathous bicolormveConte soe ec nena. tee se NA ha baie Seca 167 Hemicrepidius bilobatus (Say) ...............- INGAR See ecrvsrsrccr aye 180 “s carbonatus (LeConte) .........- INA a Sev Sinrs sie: 180 n sp., near carbonaitus (LeConte)..N.A,, ....-5..:...-- 180 vi hemipodus (Say) .2.60c22s30 +0 IN Ais Bulan 180 . memnonius (Herbst) .........-- NAVA eae nen 180 Hypnoidus dubtus (Horn)! ............-2.2+65- INAS on he Sener ieee ere 187 . Teas (Gee socoaosocapoeccnoc Buty Ae ee 187 rf musculus (Eschscholtz) ........... NGAn gat rootecit cae 187 Lepturoides denticornis (Kirby) .............. INGA Sain eeracn cn serahiecs 169 . julous (Motschulsky) ........:....0. IN At Deis cee ray tenors 169 oe lmeorism G@ainnaeus)) sere c ase Uns ism ayae scan aks 169 Remontussaeger eGonte’.......0% .<6s2 esses ose ne INGAR ase ctesne: ceisrncters 163 “_ aeneoniger (DeGeer) ............... Ui eee se ieckesaoee 159 PMONCO PSI CCONE v.26 a ie eb oe cist on as ING AY caste aa Suse 164 * californicus (Mannerheim) .......... NiAge saat: yasmin 164 een C OM US@IEG CONTEy a ircratecr ster ah ree ern IN EAS sirsten. Sacerstois apse 164 me COMpususmlbe CONnter sam at semmlc nee cise INA aera ech oeeieye 162 oO CORSE HERS SNEWS” Sooccascdcosuecoce INEAD. oe aare ewe eioarans 159 1 GOT IS ALAC. Gong ssceseoonbencas ING AR ee crepe ante yaa 164 BN ECIN DIS WC Say,)s © ar eccic. ata eee aes, INCA Sr eis craresd ae nears 164 OP ESP MCAT SECLY PSUS. ICOAY )is oisiass on oe evar NAG. ce Movers ack noose 164 =" Soeciaentahs: CandeZzes «2/22 i... « Somplicolls (Germat,)\ 24.2% 2 ass ae ete Burs eee 64 * i appressusi(Randall), 2255 minder acre NGAGE EEE 2 eee 135 x * appropinquans (Randall) says eer ee NBAs SAE eee 44 # “ Ibipustulatus, ((lsinnaeus)) \ 4:2). ee ee | Do] aR ea Rt 106 # S ontonnennns ((GSsweN) soocconcccncebadt INisAs.) Rah ere QI * ef Gastaneuse leimnaeus) ieee eee erect Bune ee tieeeerere 89 * = eiiouns (ieeheiD) Goncanoobaps.ocodcooce Bits eee 104 fo) cruciatus festivus (LeConte)........... IN| AChE sre) Sea 62 * 2 CNS (BeOS) Gonocbancboeuadoot ur.) Ose Sesh vere area 126 # ‘“ cupreus aeruginosus (Fabricius)........ ouits oie 127 x cylindriformis (Elerbst) ........0-022- IN AN Ee Ceoc tater 156 * divaricatus “((LeCGonte)))j4....42-2 eee eee INGA. SAeS ae etree 118 = S sglaucus @Germar) ieoeso. eee enn INRA. 2 a ieee 50 to =) WMeroglypiicus (Say)! assassin eerie NVA: of ee abe eee 73 ) ntlatus i(Saiyayiic. acres noe act ete ee INVAY ah 55 af ng Rendall Giirby) paee ten aera eee INGAR® 5 sticieee eee 134 x lors (Qe xa) Boesooonesdbocsoondos | Eo | Paar enor eCI Is 'c.c 48 BK Sl limoninfonmes uCelorn) seater IN GAG sons, atintereveu terete I51 * a noceonous (ESsmeI)) Soeboaccesocoondoos INGAGS 1), cuca acunieroe renee 92 (G&) nwelancholicisa(Habricitis) manaaacees eee Eur. sea 64 ae e Tonosiacmsis (ORs) pesncconscoodcsooc NAG eee 81 oO - rungire@rnes (ABWWASE)) pokcocaosdcoonsooose But: vce 118 * 2 roinenlinig (QUACCIWS)) Gaenacncouenseadods NAG Seok oe eee 112 # sn PeEtiniconnism CleinnaAcus) maemteeaeeee tie Eutey .ccohhse eee 131 * “ _ propola propola (LeConte).....5.....2- N.A® os Se eee 68 “ eS iaNens (UACOMS)) Bossccbos cdoubasec INGAG) ouSidadareneeeere 143 * SS TpruaninsaGHornm)) i ware ee ee oe INGAE 3.2 45 # SP PUG ENS TOWNE wav chine tout ate eee N. Atco. Soe ee 74 CE). AS shurpuncusAGe oda) caceusees aera mee Bar s0h..iaee eee 03 # SPN ENROS MCEleEDS ta) ten she cee ee eee a NVA AS ate 142 * “_ resplendens aerarius (Randall)......... N Aw 2 aoe 136 # ~ rotundicolliss(Say,))\(-Basterm Species )) 2 NAG) ae ee ee eeeene 96 # wu rotundicollis (Say) G Western Species). NvAW js... eee 07 oO “Gono micniadlhs IAW sooouaonssusoodcoKoK6 NAS eae ee ease 117 # ae SSeMUNAIUS. CSAW)\ Ate ener e NAS 22a eee 66 x tt SCXMOMSMBEOWIM laden eee ree NiAg 205 2 eee 57 # “* “sjaclandicus |CMallen)! 2s sey as eee Bur. & NvAG eae 142 (x) > spretuess (Nianieraeim))saneerceeeeeice cate DIS aoc 3 63 # ©.) stleicollis (GSay) (sc. te oe eee NAG ee. 5e ole ae 102 x tesscliatuss (einnaens) ane ee eee Bute V2 Goer 84 * My iraundulatism@nandall) ieee eee Ni Ag 33s 2. 76 x 4M ZH OUIES STO Waly ee eee eT aE INGAM es ccia eee 90 GE): "sirens (Schranley cae eee eee Bure) tic. cm «2c 133 d Melanactes densus'WeContes) .) i: 5.0 2aee nee INBAG, [ois Oe ae 189 NO} Et LARVAE OF THE ELATERID BEETLES—GLEN 199 LITERATURE CITED ALTUM, BERNHARD. 1878. Die forstschadlichen Elatern. Dankelm Zeitschr., vol. 10, pp. 73-81. ANDERSON, W. H. 1936. A comparative study of the labium of coleopterous larvae. Smith- sonian Misc. Coll., vol. 95, No. 13, pp. 1-20, 8 pls. ARNASON, A. P. 1931. A morphological study of the immature stages of Cryptohypnus noc- turnus Eschscholtz and a study of some ecological factors concern- ing wireworms. Report to the Canadian Division of Entomology and to the University of Saskatchewan, 105 typewritten pages, illus. (M.Sc. thesis. ) Bei-BreEnxko, G. Ya. 1928. Insects injuring root crops in the Omsk region. (In Russian.) Trud. Siber. Inst. Sel. Khoz. Lesovod, vol. 10, pp. 275-208. (Rev. Appl. Ent., vol. 17A, p. 200.) Bei-BienxKo, G. Ya., Gricor’EvA, T. G., and CHETUIRKINA, I. A. 1936. Characters of the surface and soil fauna in the steppe biocenoses near Saverovka Village, Orenburg Province. Jn Summary of the scien- tific research work of the Institute of Plant Protection for the year 1935. (In Russian.) Leningrad Acad. Sci., pp. 78-82. (Rev. Appl. Ent., vol. 24A, p. 141.) BELING, TH. 1883-1884. Beitrag zur Metamorphose der Kaferfamilie der Elateriden. Deutsche Ent. Zeitschr., vol. 27, pp. 129-144, 257-304, 1883; vol. 28, pp. 177-216, 1884. BLACKWELDER, RICHARD E. 1936. Morphology of the coleopterous family Staphylinidae. Smithsonian Misc. Coll., vol. 94, No. 13, pp. I-102. BLATCHLEY, W. A. 1910. The Coleoptera of Indiana. Bull. Indiana Dep. Geol. and Nat. Res., 1,386 pp., illus. BLuUNCK, H. 1925. Biologische Unterschiede schadlicher Drahtwurmarten. Nach. Deutschen Pflanzenschutzdienst, vol. 5, No. 5, pp. 37-39. (Rev. Appl. Ent., vol. 13A, p. 318.) Bovine, A. G. 1910. Nye Bidrag til Carabernes Udviklingshistorie, [ Larver til Calathus, Olisthopus, Oodes, og Blethisa. Ent. Medd., ser. 2, vol. 3, pp. 319-376. 1914. On the abdominal structure of certain beetle larvae of the campo- deiform type. Proc. Ent. Soc. Washington, vol. 16, pp. 55-61, pls. 3-6. Bovine, A. G., and CRAIGHEAD, F. C. 1931. An illustrated synopsis of the principal larval forms of the order Coleoptera. Brooklyn Ent. Soc. Publ., 351 pp. Brown, W. J. 1930. A revision of the North American species of Eanus. Canadian Ent., _ vol. 62, pp. 161-166. 1935a. American species of Ludius; the cruciatus and edwardsi groups (Coleop.). Canadian Ent., vol. 67, pp. 1-8. 200 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III 1935b. American species of Ludius; the aeripenmis group. Canadian Ent., vol. 67, pp. 125-135. 1936a. American species of Ludius; the semivittatus and nitidulus groups (Coleoptera). Canadian Ent., vol. 68, pp. 11-20. 1936b. American species of Ludius; the fallax and triundulatus groups. Canadian Ent., vol. 68, pp. 99-107. 1936c. American species of Ludius; the inflatus group. Canadian Ent., vol. 68, pp. 133-136. 1936d. American species of Ludius; the propola group. Canadian Ent., vol. 68, pp. 177-187. 1939. Some American species of Ludius (Coleoptera). Canadian Ent., vol. 71, pp. 44-49. CHAPUIS, F., and CANDEZE, E. 1855. Catalogue des larves des Coleopteres. Mem. Soc. Roy. Sci. Liége, vol. 8, pp. 139-148, 3 figs. CHRZANOWSKI, A. 1931. Beitrag zur Kenntnis der von Drahwiirmern-Elateridae-Larven her- beigefirrten Beschadigungen von Kulturpflanzen. Choroby Roslin, vol. 1, pp. 51-52. Warsaw. (Rev. Appl. Ent., vol. 22A, pp. 437-438.) Comstock, J. H., and SLINGERLAND, M. V. 1891. Wireworms. New York (Cornell) Agr. Exper. Stat. Bull. 33, PP. 191-272, 21 figs. Davis, J. J. 1911. Preliminary report on the more important insects of the truck gardens of Illinois. Jn Illinois Farmers’ Inst. 16th Ann. Rep., p. 251. DEVEREUX, W. L. 1878. Tetraopes tetrophthalmus Forst. Canadian Ent., vol. 10, p. 143. pu Buysson, H. 1888. Note sur les larves d’Elaterides. Rev. d’Ent., vol. 7, pp. 14-17. Durnovo, Z. P. 1935. The wireworm Corymbites sjaelandicus Mill. as a pest of vegetable garden plants. (In Russian.) Plant Protect., fasc. 1, p. 150. Lenin- grad. (Rev. Appl. Ent., vol. 23A, p. 573.) Essig) Ha ©} 1926. Insects of western North America, 1,035 pp., 766 figs. Macmillan Co., New York. Evans, W. 1921. Cryptohypnus riparius, a click-beetle as a possible agricultural pest. Scottish Nat., Nos. 119-120, pp. 181-182. (Rev. Appl. Ent., vol. 10A, p. 177.) Ferris, G. F. 1934. Setae. Canadian Ent., vol. 66, pp. 145-150. Fitcu, ASA. 1867. Wireworms. Jn Eleventh Report on the Insects of New York State for 1866. Trans. New York State Agr. Soc., pp. 519-543, illus. Forses, S. A. 1892. Wireworms. In Eighteenth Report of the State Entomologist on the Noxious and Beneficial Insects of Illinois, pp. 24-44. (Reprinted 1920. ) NO. Tt LARVAE OF THE ELATERID BEETLES—GLEN 201 Forp, G. H. 1917. Observations on the larval and pupal stages of Agriotes obscurus L. Ann. Appl. Biol., vol. 3, pp. 97-115. GuHILAroy, M. S. 1937. The fauna of injurious soil insects of arable land. Bull. Ent. Res., vol. 28, pt. 4, pp. 633-637. GLEN, ROBERT. 1931. The external morphology and the characters separating the elaterid larvae of Saskatchewan. Report to the Canadian Division of Ento- mology and to the University of Saskatchewan, 118 typewritten pages, illus. (M.Sc. thesis.) 1935. Contributions to the morphology of the larval Elateridae (Coleop- tera) ; No. 1—Ludius aeripennis destructor Brown. Canadian Ent., vol. 67, pp. 231-238, 2 pls. 1941. Contributions to the morphology of the larval Elateridae (Coleop- tera); No. 2—Agriotes limosus LeConte. Canadian Ent., vol. 73, No. 4, pp. 57-62, 2 pls. GLEN, Rosert, and Kine, K. M. 1938. The relation of wireworms to potato growing in the Prairie Prov- inces. Sci. Agr., vol. 18, pp. 283-287. Guten, Rosert, Kine, K. M., and Arnason, A. P. 1943. The identification of wireworms of economic importance in Canada. Canadian Journ. Res., vol. 21, sect. D, pp. 358-387, 64 figs. GRAF, JOHN E. 1914. A preliminary report of the sugar-beet wireworm. U. S. Dep. Agr., Bur. Ent. Bull. 123, 68 pp., 9 figs., 23 pls. GUENIAT, EpMOND. 1934. Contribution a l’étude du developpement et de la morphologie de quelques Elaterides. Mitt. Schweiz. Ent. Ges., vol. 16, pp. 167-298, 30 figs. (These, Ecole Polytechn. Fed. Zurich, 133 pp., illus.) Hawkins, J. H. 1936. The bionomics and control of wireworms in Maine. Maine Agr. Exper. Stat. Bull. 381, 146 pp., illus. HENRIKSEN, K. L. 1911. Oversigt over de danske Elateride larvar. Entom. Medd., vol. 4, pp. 225-331, illus. Hopkins, A. D. 1909. Contributions toward a monograph of the scolytid beetles. 1. The genus Dendroctonus. U. S. Dep. Agr., Bur. Ent. Bull. 17 (Techn. Ser.), pt. I, pp. 57-64. Horst, ALBERT. 1922. Zur Kenntnis der Biologie und Morphologie einiger Elateriden und ihrer Larven (Insbesondere Untersuchungen iiber Agriotes ob- scurus L.). Arch. Naturg., vol. 88, pp. 1-90, illus. Hys op, J. A. 1915a. Wireworms attacking cereal and forage crops. U. S. Dep. Agr. Bull. 156, 34 pp., illus. 1915b. Notes on the habits and anatomy of Horistonotus uhleru Horn. Proc. Ent. Soc. Washington, vol. 17, pp. 179-185, illus. 202 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. DIE 1917. The phylogeny of the Elateridae based on larval characters. Ann. Ent. Soc. America, vol. 10, pp. 241-263, illus. 1921. Genotypes of the elaterid beetles of the world. Proc. U. S. Nat. Mus., vol. 58, pp. 621-680. Hystop, J. A., and Bovine, A. G. 1935. Larva of Tetrigus fleutiauxi Van Zwaluwenburg. Proc. Hawaiian Ent. Soc., vol. 9, pp. 49-61, illus. Jewett, H. H. 1939. Description of the larva of Ludius divaricatus Lec. Canadian Ent., vol. 71, pp. 105-108, illus. KINCAID, TREVOR. 1900. The metamorphoses of some Alaska Coleoptera. In Papers from the Harriman Alaska Expedition. Proc. Washington Acad. Sci., vol. 2, pp. 367-388, illus. Kine, K. M. 1928. Economic importance of wireworms and false wireworms in Sas- katchewan. Sci. Agr., vol. 8, pp. 693-706. Kine, K. M., Arnason, A. P., and GLEN, R. 1933. The wireworm problem in field crops of western Canada: A dis- cussion of eleven. years’ results. Canadian Dep. Agr., Saskatoon Ent. Lab. Leaflet No. 35, 21 numbered leaves. Kinc, K. M., Giten, R., McManon, H., and Arnason, A. P. 1940. Wireworm control in western grain fields. Canadian Dep. Agr., War- time Prod. Ser. Special Pamphlet No. 37, 4 pp. KNULL, J. N. 1930. Notes on Coleoptera—No. 2. Ent. News, vol. 41, p. 83. 1932. Notes on Coleoptera—No. 3. Ent. News, vol. 43, p. 43. 1934. Notes on Coleoptera—No. 4. Ent. News, vol. 45, p. 208. KorscHELt, E. 1924. Der Gelbrand Dytiscus marginalis L., vol. 2, 964 pp., 405 figs. LANCHESTER, H. P. 1939. The external anatomy of the larva of the Pacific coast wireworm. U. S. Dep. Agr. Techn. Bull. 693, 44 pp., illus. 1941. The larva of Limonius consimilis Walk. Ann. Ent. Soc. America, vol. 34, pp. 367-376, illus. LANE, M. C. 1925. The economic wireworms of the Pacific Northwest (Elateridae). Journ. Econ. Ent., vol. 18, pp. 90-08. 1931. The Great Basin wireworm in the Pacific Northwest. U. S. Dep. Agr. Farmers Bull. 1657, 8 pp., illus. 1935. Recent progress in the control of wireworms. Proc. World’s Grain Exhib. and Conf. 1933, vol. 2, pp. 529-534. (Rev. Appl. Ent., vol. 23A, p. 539.) 1941. Wireworms and their control on irrigated lands. U. S. Dep. Agr. Farmers Bull. 1866, 21 pp., illus. LANGENBUCH, R. 1932. Beitrage zur Kenntnis der Biologie von Agriotes obscurus L. und Agriotes lineatus L. Zeitschr. Ang. Ent., vol. 19, pp. 278-300. LENG, C. W. 1920. Catalogue of the Coleoptera of America, north of Mexico, 470 pp. John D. Sherman, Jr., Mt. Vernon, N. Y. NOS ET LARVAE OF THE ELATERID BEETLES—GLEN 203 LINNANIEMI, W. M. 1935. Kertomus tuhoelainten essiintymisesta suomessa vuosina I917-1923. Valt. Maatalousk. Julk No. 68, 159 pp. (Rev. Appl. Ent., vol. 23A, pp. 516-517.) MasairTis, A. I. 1929. Data on the fauna and biology of elaterids in Siberia. (In Russian.) Izv. Sibirsk. Kraev. Stantz. Zascch. Rast., No. 3 (6), pp. I-41. Tomsk. (Rev. Appl. Ent., vol. 18A, p. 48.) 1931. On the morphology of the larva of Selatosomus spretus Mannh. (In Russian with English summary.) Plant Protect., vol. 8, pp. 203-207, illus. McDouea.i, W. A. 1934. The determination of larval instars and stadia of some wireworms (Elateridae). Queensland Agr. Journ., vol. 42, pp. 43-70, illus. MEQuIGNON, A. 1930. Notes synonymiques sur les Elaterides (Col.) (4th note). Bull. Ent. Soc. France, No. 4, pp. 91-96. Munro, J. A., and Scu1F1Ino, Louts A. 1938. Preliminary studies of wireworms affecting potato tubers in North Dakota. .Journ. Econ. Ent., vol. 31, pp. 487-488. NORDLINGER, H. 1880. Lebensweise von Forstkerfen oder Nachtrage zu Ratzeburgs Forst- insekten. Zweite Vermehrte Auflage, 73 pp. Stuttgart. Perris, EDOUARD. 1863. Histoire des insectes du pin maritimes, vol. 1, Coleopteres. Paris. 1877. Larves des Coleopteres, pp. 161-188. Paris, Deyrolle, Naturaliste. Pityucina, S. A. 1937. The principal conclusions from a study of wireworms in the depart- ment of West Kazakstan. (In Russian.) Jn Summaries of the work of the Section of Plant Protection of the All-Union Institute of Grain Farming (Saratov) for the year 1936. Plant Protect., vol. 13, pp. 95-96. (Rev. Appl. Ent., vol. 26A, p. 76.) PosPELova, V. 1937. Study of wireworms of Tomsk district, West Siberian Region. (In Russian.) J Summary of the scientific research work of the Institute of Plant Protection for the year 1936. Pt. 1, Pests and diseases of cereals and shelterbelts. Leningrad Acad. Agr. Sci., pp. 46-48. (Rev. Appl. Ent., vol. 26A, pp. 425-426.) RAMBOUSEK, F. 1928. Uber die Felddrahtwiirmer. I. Systematischer Teil. Zeitschr. Zucker- ind. Czech. Repub., vol. 52, pp. 393-402, illus. Prague. (Rev. Appl. Ent., vol. 17A, p. 208.) 1929. Uber die Felddrahtwirmer. II. Biologischer Teil. Zeitschr. Zucker- ind. Czech. Repub., vol. 54, pp. 197-201. (Rev. Appl. Ent., vol. 184, p. 205.) REGNIER, R. 1921. Un ennemi des plantes protageres, Corymbites (Diacanthus) latus (Elaterides). Bull. Soc. Path. Veg. France, vol. 8, pp. 21-24. (Rev. Appl. Ent., vol. 9A, p. 402.) 14 204. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE. nick 1932. Renseignements complementares sur la biologie de Selatosomus (Co- rymbites) latus F. (Elateridae). Bull. Soc. Sci. Nat. Rouen, vols. 66-67 (1930-1931), pp. 13-15. 1Ptong (C- 1887. Essai d’études sur certaines larves de Coleopteres. Ann. Linn. Soc. Lyon, pp. 199-208. Ripeway, R. 1912. Color standards and nomenclature. Roserts, A. W. R. 1919-1928. On the life-history of “wireworms” of the genus Agriotes Esch., with some notes on that of Athous haemorrhoidalis F. Pts. 1-4. Ann. Appl. Biol., vol. 6, pp. 116-135, illus., 1919; vol. 8, pp. 193-215, illus., 1921; vol. 9, pp. 306-324, illus., 1922; vol. 15, pp. 90-94, illus., 1928. 1930. A key to the principal families of Coleoptera in the larval stage. Bull. Ent. Res., vol. 21, pt. I, pp. 57-72, illus. RUPERTSBERGER, MATHIAS. 1870. Biologische Beobachtungen, Coleoptera. Verhandl. Zool. Bot. Ges. Wien, vol. 20, pp. 835-842. 1880. Biologie der Kafer Europas. Linz a. d. Donau, pp. 159-163. 1894. Die biologische Literatur iiber die Kafer Europas von 1880 an. Linz a. d. Donau und Niederrana, pp. 163-168. Rye, BERTRAM G. 1906. Fortegnelse over Danmarks Biller, 166 pp. Copenhagen. SAALAS, UUNIO. 1923a. Studien iiber die Elateriden Finnlands I Corymbites cupreus Fabr. subsp. aeruginosus Fabr. und seine Verheerungen, besonders in der Gegend von Kainuu. Ann. Soc. Zool.-Bot. Fennicae Vanamo, vol. 2, pp. 121-168. 1923b. Die Fichtenkafer Finnlands II Spezieller Teil 2 und Larvenbestim- mungstabelle, 746 pp., 28 pls. Helsinki. 1937. Ein neuer Larventyp der Elateriden. Mutmasslich Orithales serrati- cormis Payk. (Col., Elateridae). Ann. Ent. Fennici, vol, 3, pp. 65- 73, illus. 1938. Berichtigung zu meinem Aufsatz “Ein neuer Larventyp der Ela- teriden.” Durfte Eanus (Paranomus) costalis Payk. und nicht Orithales serraticornis Payk. (Col., Elateridae) sein. Ann. Ent. Fennici, vol. 4, pp. 53-55. SCHENKLING, S. 1927. Elateridae. Jn Coleopterorum Catalogus (ed. Junk), vol. 2, pt. 88. ScuHIoptTE, J. C. 1870. De Metamorphosi Eleutheratorum Observationes. Naturh. Tiddsk., vol. 6, pp. 467-536, pls. 3-10. SEMENOv, A. E. 1931. Pests of Makhorka tobacco. (In Russian.) Med., 32 pp., 29 figs., 5 refs. Moscow, Gosud. Tekhn. Izd. (Rev. Appl. Ent., vol. 19A, p. 502.) SLATER, J. W. 1869. Habitats of Ctenicerus pectinicornis and C. cupreus. Ent. Month. Mag., vol. 5, p. 276. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 205 Snoperass, R. E. 1931. Morphology of the insect abdomen, pt. 1. Smithsonian Misc. Coll., vol. 85, No. 6, 128 pp. 1935. Principles of insect morphology, 667 pp. Stone, M. W. 1941. Life history of the sugar-beet wireworm in southern California. U. S. Dep. Agr. Techn. Bull. 744, 87 pp., illus. STRICKLAND, E. H. 1926. Wireworms of Alberta. Univ. Alberta Coll. Agr. Res., Bull. No. 2, 18 pp., illus. 1935. The biology of prairie inhabiting wireworms. Proc. World’s Grain Exhib. Conf. Regina, Canada, 1933, vol. 2, pp. 520-529. 1938. Insect pests of grain in Alberta. Univ. Alberta Coll. Agr. Bull. No. 24 (revised), pp. 33-40. 1939. Life cycle and food requirements of the northern grain wireworm, Ludius aeripennis destructor Brown. Journ. Econ. Ent., vol. 32, PP. 322-320. 1942. Variations in the length of the life-cycle of wireworms. Journ. Econ. Ent., vol. 35, pp. 109-110. SUBKLEwW, W. 1934a. Die Bekampfung der Drahtwiirmer: Nachrbl. Deutsch. Pfl. Sch. Dienst., vol. 14, pp. 52-53. (Rev. Appl. Ent., vol. 22A, p. 386.) 1934b. Agriotes lineatus L. und Agriotes obscurus L. (Ein Beitrag zu ihrer Morphologie und Biologie.) Zeitschr. Ang. Ent., vol. 21, pp. 96- 122, illus. (Rev. Appl. Ent., vol. 22A, p. 466.) Tuomas, C. A. 1930. A review of research on the control of wireworms. Pennsylvania State Coll. Exper. Stat. Bull. 259, 52 pp. 1940. The biology and control of wireworms. Pennsylvania State Coll. Exper. Stat. Bull. 392, 90 pp. 1941. The Elateridae of Pennsylvania. Journ. New York Ent. Soc., vol. 40, Pp. 233-263. TorrE-BuENoO, J. R. DE LA. 1937. A glossary of entomology. Brooklyn Ent. Soc. Publ., 336 pp., illus. VAN Dyke, EpwIn C. 1932. Miscellaneous studies in the Elateridae and related families of Cole- optera. Proc. California Acad. Sci., 4th ser., vol. 20, pp. 291-465. VAN ZWALUWENBURG, R. H. 1939. Larvae of Hawaiian elaterid beetles. Proc. Hawaiian Ent. Soc., vol. 10, pp. 275-279, illus. WEEsE, A. O. 1924. Animal ecology of an Illinois elm-maple forest. Illinois Biol. Monogr., vol. 9, p. 8. WEst, AUGUST. 1937. Tillaeg og Rettelser til Fortegnelserne over de danske Coleoptera III (sluttet). Entom. Medd., p. 484. WHITEHEAD, W. E. 1932. The morphology of the head-capsule of some coleopterous larvae. Canadian Journ. Res., vol. 6, pp. 227-252. 206 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III XAMBEU, LE CAPITAINE. 1895-1896. Moeurs et metamorphoses des insectes, 5 Mem. Ann. Soc. Linn. Lyon, vol. 42, pp. 87-91. 1912-1914. Moeurs et metamorphoses des insectes, 18 Mem. Ann. Soc. Linn. Lyon, vol. 59, pp. I11-161, 1912; vol. 60, pp. 1-34, 123-146, 1914. ZNAMENSKY, A. V. 1926. Insects injurious to agriculture. Part 1, Pests of grain crops. (In Russian.) Trans. Poltava Agr. Exper. Stat., No. 50, Ent. Div. No. 13, 206 pp., 7 pls., 118 figs. (Rev. Appl. Ent., vol. 14A, p. 510.) 1927. Instructions for carrying out observations on the entomofauna of the soil. (In Russian.) Trans. Poltava Agr. Exper. Stat., No. 51, Ent. Div. No. 14, 58 pp., 72 figs. (Rev. Appl. Ent., vol. 15A, p. 314.) NO. IT LARVAE OF THE ELATERID BEETLES—GLEN 207 epla— _ Fic. 1.—Ludius aeripennis destructor Brown. a, head, dorsal view; b, head, ventral view. 208 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 2.—Ludius aeripennis destructor Brown. a, third segment and tip of second segment of left antenna, mediodorsal view; b, whole larva, dorsal view; c, nasale and subnasale, ventral view; d, left antenna, ventral view; e, right mandible, dorsal view; f, right mandible, medial view ; g, head, lateral view; h, right mandible, ventral view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 209 \ N hphsc- -S \ x Fic. 3.—Ludius aeripennis destructor Brown. a, tip of maxillary palpus, dorsal view (oil immersion) ; b, maxillary palpus, medioventral view; c, right maxilla, medioventral view; d, tip of galea, dorsal view (oil immersion) ; e, right galea, medioventral view; f, preoral cavity with hypopharynx in situ, dorsal view; g, posterior portion of hypopharynx, dorsal view; h, “sensory” peg from dorsal surface of ligula; 7, first prementum and sec- ond prementum, ventral view; 7, tip of labial palpus, dorsal view (oil immer- sion); k, first prementum, dorsal view. 210 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. ITI x py ug ia Hi i i Mh \ a a We mld vad at! A Fic. 4—Ludius aeripennis destructor Brown. a, tentorium, in situ, dorsal view (semidiagrammatic) ; b, right half of ten- torium, lateral view (semidiagrammatic) ; c, branching filaments from hypo- pharynx; d, dorsal aspect of preoral cavity; e, “sensory” plate and surrounding sensilla on subnasal flap (oil immersion) ; f, left half of “subnasal flap.” NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 211 tg mtg trim loim i / \ / par 2 / \ / \ \ i) \ Vs \ | yt oN \ \ v4 Ue | N \ \ “| A 1 prst = epst epm prst Itg Itg> epst eust epm Fic. 5.—Ludius acripennis destructor Brown. a, thorax, lateral view; b, thorax, ventral view (left legs removed; setae omitted from metathorax) ; c, episternum of mesothorax, lateral view, to show spinelike setae; d, left leg of mesothorax, anterior aspect. 212 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 6.—Ludius aeripennis destructor Brown. a, fourth abdominal segment, lateral view; b, fourth abdominal segment, ven- tral view; c, mediotergite of fourth abdominal segment, dorsal view; d, ninth and tenth abdominal segments, ventral view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 213 C Fic. 7.—Ludius aeripennis destructor Brown. a, ninth abdominal seg c, left urogomphus, latera view. ment, dorsal view; b, left urogomphus, dorsal view; 1 view; d, ninth and tenth abdominal segments, lateral 214 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III eee j } Fic. 8.—Representatives of subfamilies. a, Cardiophorus sp., larva, dorsal view ; subfamily Cardiophorinae. b, Dalopius parvulus Brown, larva, dorsal view; subfamily Elaterinae. c, Oestodes tenuicollis (Randall), larva, dorsal view; subfamily Oestodinae. d, Ludius tessellatus (Linnaeus), larva, dorsal view ; subfamily Pyrophorinae. NOP TT LARVAE OF THE ELATERID BEETLES—GLEN 215 J Fic. 9.—Acolus, Hypnoidus, Limonius, and Ludius. a-e, i, Aeolus mellillus marginicollis (Horn): a, frontoclypeal region, dorsal view; b, nasale and subnasale, ventral view; c, ventral mouthparts, ventral view; d, left mandible, ventral view; ¢, left mandible, dorsal view; 7, ninth and tenth abdominal segments, lateral view. f, Hypnoidus dubius (Horn): right mandible, dorsal view. g, Limonius aeneoniger (DeGeer): right mandible, dorsal view. h, j, Ludius aeripennis destructor Brown: h, ventral mouthparts, ventral view; /, frontoclypeal region, dorsal view. 216 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT —S SS ---ltgI <> Fic. 10—Eanus, Hypnoidus, Ludius, Limonius, Crepidomenus and Cryptohypnus. a, c, Eanus decoratus (Mannerheim): a, urogomphi, dorsal view; ¢, third abdominal segment, lateral view. b, Hypnoidus dubius (Horn): urogomphi, dorsal view. d, Ludius aeripennis destructor Brown: presternal area of pro- thorax, ventral view. e, Limonius dubitans LeConte: presternal area of pro- thorax, ventral view. f, Crepidomenus queenslandicus Blair: ninth abdominal segment, dorsal view. g, Cryptohypnus nocturnus (Eschscholtz) : head, thorax, and first abdominal segment, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 217 Fic. 11.—Limonius, Ludius, and Elathous. a, Limonius aeneoniger (DeGeer): urogomphi, dorsal view. 06, f, Ludius resplendens aerarius (Randall): 6, left urogomphus, lateral view; f, ninth abdominal segment, dorsal view. c, Elathous bicolor (LeConte): left uro- gomphus, lateral view. d, Limonius aeger Leconte: left urogomphus, lateral view. e, Ludius pyrrhos (Herbst): spiracle of eighth abdominal segment. g, Limonius ectypus Say (?): ninth abdominal segment, dorsal view. h, Ludius limontformis (Horn) (?): ninth abdominal segment, dorsal view. 218 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE: Eri Fic. 12.—Athous, Lepturoides, Hemicrepidius, and Ludwus. a, Athous cucullatus (Say), urogomphi, dorsal view. b, Lepturoides linearis (Linnaeus), urogomphi, dorsal view. c, Athous mutilatus Rosenhauer, uro- gomphi, dorsal view. d, Hemicrepidius sp., mediotergite of fourth abdominal segment, lateral view. ¢, Ludius divaricatus (LeConte), ninth abdominal seg- ment, dorsal view. f, Hemicrepidius memnonius (Herbst), ninth abdominal seg- ment, dorsal view (drawn from a larval exuvium). NO. II LARVAE OF THE ELATERID BEETLES—GLEN 219 opr opr d f ipr e/. / é An: ipr Nak: ® & y . tub Gi _-opr Fic. 13.—Species of Ludius. a, Ludius nitidulus (LeConte): presternal area of prothorax, ventral view. b, L. resplendens aerarius (Randall): presternal area of prothorax, ventral view. c, L. propola propola (LeConte): left urogomphus, lateral view. : d, L. aeripennis destructor Brown: left urogomphus, lateral view. e, L. sexualts Brown (?): left urogomphus, lateral view. f, 1, L. glaucus (Germar) : f, left urogomphus, lateral view; h, urogomphi, dorsal view. g, L. triundulatus (Ran- dall) : left urogomphus, lateral view. i, L. pruininus (Horn): urogomphi, dorsal view. 220 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 14.—Species of Ludius. a, b, Ludius appressus (Randall): a, left urogomphus, lateral view; b, uro- gomphi, dorsal view. c, L. bipustulatus (Linnaeus): urogomphi, dorsal view. d, L. triundulatus (Randall) : mediotergite of fifth abdominal segment, lateral view. e, L. aeripennis destructor Brown: mediotergite of fourth abdominal seg- ment, lateral view. f, L. sjaelandicus (Miller): eighth abdominal segment, lateral view. g, L. kendalli (Kirby): urogomphi, dorsal view. h, L. cinctus (Paykull): left maxilla, ventral view. 7, L. nitidulus (LeConte): ninth abdominal segment, dorsal view. j, L. rotundicollis (Say) (“Western species”), ninth abdominal segment, dorsal view. bo iS) — NOT EE LARVAE OF THE ELATERID BEETLES—GLEN Fic. 15.—Ludius aeripennis group. a-c, Ludius pruininus (Horn): a, ninth abdominal segment, dorsal view ; b, left urogomphus, lateral view; c, left urogomphus, dorsal view. d-f, L. latus (Fabricius) : d, left urogomphus, lateral view; e, left urogomphus, dorsal view ; f, ninth abdominal segment, dorsal view. 222 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. ‘ITI e Fic. 16.—Species of Ludius. a, Ludius glaucus (Germar): ninth abdominal segment, dorsal view. b, e, L. semivittatus (Say) (?): b, first abdominal segment, ventral view (semi- diagrammatic, the larva being much shrunken); ¢, left urogomphus, lateral view. c, L. cruciatus festivus (LeConte) (?): left urogomphus, lateral view. d, f, L. sexualis Brown (?): d, left urogomphus, lateral view; f, ninth abdomi- nal segment, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 223 Fic. 17.—Ludius propola group. a-c, Ludius propola propola (LeConte): a, ninth abdominal segment, dorsal view; b, right antenna, medial view; c, left urogomphus, lateral view. d, e, L. pudicus Brown: d, urogomphi, dorsal view; e, left urogomphus, lateral view. f, g, L. hieroglyphicus (Say): f, urogomphi, dorsal view; g, left urogomphus, lateral view. 224 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 18.—Ludius triundulatus group. a-d, f, g, Ludius triundulatus (Randall): a, nasale and subnasale, ventral view ; b, mediotergite of fifth abdominal segment, dorsal view; c, fifth abdominal segment, ventral view; d, second and third segments of left antenna, ventral view ; f, left urogomphus, lateral view; g, ninth abdominal segment, dorsal view. e, L. nebraskensis (Bland) (?): left urogomphus, lateral view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN to iS) on 4) { i 1 | 4 ' I as ee ee ee Fic. 19.—Ludius fallax group. a, b, d, f, g, Ludius tessellatus (Linnaeus) : a, head, lateral view; b, frontocly- peal region, dorsal view; d, third segment and tip of second segment of left antenna, medioventral view; f, gular area, showing position of tentorium, ventral view; g, mediotergite of fourth abdominal segment, dorsolateral view. c, e, L. castaneus (Linnaeus): c, nasale and subnasale, ventral view; e, third seg- ment and tip of second segment of left antenna, lateral view. 226 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL DLT S Fic. 20.—Ludius fallax group. a, b, e, Ludius tessellatus (Linnaeus): a, ninth abdominal segment, dorsal view; b, ninth and tenth abdominal segments, lateral view; e, left urogomphus, lateral view. c, L. castaneus (Linnaeus) : urogomphi, dorsal view. d, f, L. bombycinus (Germar): d, urogomphi, dorsal view (drawn from larval exu- vium) ; f, left urogomphus, lateral view (drawn from larval exuvium). eee NO. II LARVAE OF THE ELATERID BEETLES—GLEN 227 lo x 8.0240 NY Abi ~ Se eon a *. 0 Cy Cue . °° 2° signe: ae, sii: a Fic. 21.—Ludius rotundicollis group. a, b, e, g, h, Ludius rotundicollis (Say) (‘Western species’) : a, head, lateral view (ventral mouthparts omitted) ; b, whole larva, dorsal view; e, mediotergite of fourth abdominal segment, dorsolateral view; g, ninth abdominal segment, dorsal view; hf, ninth and tenth abdominal segments, lateral view. c, d, L. sulcicollis (Say): c, left urogomphus, dorsal view; d, left urogomphus, lateral view. f, L. rotundicollis (Say) (“Eastern species”) : left urogomphus, lateral view. 228 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 22.—Species of Ludius. a-c, e, Ludius cinctus (Paykull) : a, whole larva, dorsal view (showing color pattern) ; b, left maxilla, ventral view; c, ninth abdominal segment, dorsal view (color pattern omitted) ; ¢, left urogomphus, lateral view. d, f, L. bipustulatus (Linnaeus) : d, ninth abdominal segment, dorsal view (color pattern omitted ) ; f, eighth abdominal segment, dorsal view (showing color pattern). ELATERID BEETLES—GLEN 229 LARVAE OF THE NO. II EO _f-—trim TN HU EEL | —_ > >}--pim F ‘ eae | ; GR :ePPley: x ee | Vas i { A\ f- . he yi oP y —f— | \ SS p Kia / AN N Se Eo + hay >to LS - : ur x ! LY 1 \ \ \ ! 7 vi H \ \ u cn jipr\ opr Fic. 23.—Ludius mitidulus group. a, b, e, g, Ludius nitidulus (LeConte) : a, frontoclypeal area, dorsal view; b, left mediotergite of seventh abdominal segment, dorsolateral view; e, left urogomphus, lateral view; g, ninth abdominal segment, dorsal view. ¢, L. rufo- d, f, L. nigricornis (Panzer) (?): pleuralis Fall (?): nasale, dorsal view. . d, nasale, dorsal view; f, left urogomphus, dorsal view. 230 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 24.—Ludius divaricatus (LeConte). a, frontoclypeal area, dorsal view; b, right mandible, dorsal view; c, ninth and tenth abdominal segments, lateral view; d, ninth abdominal segment, dorsal view. INGOs EE LARVAE OF THE ELATERID BEETLES—GLEN 231 g Fic. 25.—Species of Ludius. a, Ludius cupreus cupreus (Fabricius) : nasale and subnasale, ventral view. b, c, f, L. appressus (Randall) (drawn from larval exuvium): b, nasale and subnasale, ventral view; c, tip of left urogomphus, lateral view; f, urogomphi, dorsal view. d, e, 9, h, L. cupreus aeruginosus (Fabricius) : d, left mediotergite of eighth abdominal segment, laterodorsal view; e, fourth abdominal segment, ventrolateral view; g, left urogomphus, lateral view; /, ninth abdominal seg- ment, dorsal view. 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III iS) ios) i) Fic. 26.—Ludius resplendens aerarius (Randall). a, nasale and subnasale, ventral view; 5, setal arrangement in left dorsal epicranial sulcus; c, left mediotergite of third abdominal segment, dorsal view ; d, ungula of right prothoracic leg, anterior aspect; e, left urogomphus, lateral view; f, ninth abdominal segment, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 233 yyy) i Fic. 27.—Ludius sjaelandicus (Miller). a, nasale and subnasale, ventral view; b, gular area, ventral view; c, ninth abdominal segment, dorsal view; d, seventh, eighth, ninth, and tenth abdominal segments, lateral view. 234 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III nsa~ N Fic. 28.—Ludius pyrrhos (Herbst). a, seventh and eighth abdominal segments, lateral view; b, left. mandible, dorsal view; c, nasale, dorsal view ; d, ninth abdominal segment, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 235 Fic. 29.—Ludius limoniiformis group. a-c, Ludius limoniiformis (Horn) (?): a, nasale and subnasale, ventral view ; b, ninth abdominal segment, dorsal view; c, left urogomphus, lateral view. d-e, L. cylindriformis (Herbst) (?): d, left urogomphus, lateral view; e, ninth abdominal segment, dorsal view. 236 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III yn ‘Tepe ee Plates ; gi EAE ES be py Fic. 30.—Species of Limonius. a-d, f, Limonius aeneoniger (DeGeer) (=Pheletes bructeri Panzer): a, frontoclypeal area, dorsal view; b, nasale and subnasale, ventral view; c, right mandible, dorsal view (probably somewhat eroded) ; d, left urogomphus, lateral view ; f, ninth abdominal segment, dorsal view. e, g, h, L. pilosus (Leske) (?): e, right mandible, dorsal view; g, nasale and subnasale, ventral view; h, ninth abdominal segment, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 237 Fic. 31.—Species of Limonius. a, e, g, Limonius pectoralis LeConte: a, nasale, dorsal view; e, ninth abdominal segment, dorsal view; g, ninth and tenth abdominal segments, lateral view. b, c, L. dubitans LeConte (= Nothodes dubitans): b, nasale and subnasale, ventral view; c, presternal area of prothorax, ventral view. d, L. aeger LeConte: left urogomphus, lateral view. f, L. subauratus LeConte (?): left urogomphus, lateral view. 238 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS rit Fic. 32.—Limonius canus group. a-e, Limonius dubitans LeConte (= Nothodes dubitans): a, frontoclypeal area, dorsal view; b, right mandible, dorsal view; c, left mediotergite of fourth abdominal segment, lateral view; d, ninth abdominal segment, dorsal view; e, left urogomphus, lateral view. #, L. ectypus (Say) (?): ninth and tenth abdominal segments, lateral view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 239 Fic. 33.—Elathous bicolor and Lepturoides linearis. a, b, Elathous bicolor (LeConte) : a, left urogomphus, lateral view ; b, ninth abdominal segment, dorsal view. c-e, Lepturoides linearis (Linnaeus) : c, left mediotergite of fourth abdominal segment, dorsolateral view; d, ninth abdominal segment, dorsal view; e, left urogomphus, lateral view. 240 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, III g Fic. 34.—Species of Athous. a, c, Athous haemorrhoidalis (Fabricius) : a, left urogomphus, lateral view ; c, urogomphi, dorsal view. b, A. vittatus (F abricius ) : urogomphi, dorsal view. d, e, A. rufifrons. ( Randall) : d, urogomphi, dorsal view; e, left mediotergite of fourth. abdominal segment, dorsolateral view. f, A. brightwelli (Kirby) : urogomphi, dorsal view. 9, h, A. mutilatus Rosenhauer : g, left urogomphus, lateral view; h, urogomphi, dorsal view. NO. II LARVAE OF THE ELATERID BEETLES—GLEN 241 ret. mA AAAAA WA AANA, aA A cay) N us A, A A HANS ath Any! Fic. 35.—Athous cucullatus group. a, c, d, Athous cucullatus (Say): a, right mandible, dorsal view; c, ninth abdominal segment, dorsal view; d, left urogomphus, lateral view. be; f, A. scapularis (Say): b, nasale and subnasale, ventral view; e, left mediotergite of fourth abdominal segment, laterodorsal view; f, ninth abdominal segment, dorsal view. 242 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III wi \l\ ih ATTA aoa _A e000] i isi { | il inn vidi C e Fic. 36.—Athous undulatus group. a-c, Athous (Harminius) undulatus (DeGeer): a, urogomphi, dorsal view; b, left urogomphus, lateral view; c, left mediotergite of fifth abdominal segment, laterodorsal view. ae was villosus (Geoffroy) : d, urogomphi, dorsal view; e, left urogomphus, lateral view. NO. II LARVAE OF THE ELATERID BEETLES—-GLEN 243 Fic. 37.—Species of Hemicrepidius and Crepidomenus. a, Hemicrepidius sp., near carbonatus (LeConte) : right mandible, dorsal view (drawn from larval exuvium of a reared specimen). b, Hemucrepidius sp. (unidentified): left mediotergite of fourth abdominal segment, dorsolateral view. c, H. memnonius (Herbst): ninth abdominal segment, dorsal view (drawn from larval exuvium of a reared specimen). d-f, Crepidomenus queenslandicus Blair: d, left mediotergite of third abdominal segment, dorso- lateral view; e, ninth and tenth abdominal segments, lateral view; f, ninth abdominal segment, dorsal view. 244 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES Dit Fic. 38.—Species of Cryptohypnus. a, c, e, g, Cryptohypnus abbreviatus (Say): a, posterior part of frontoclypeal area showing adjacent setae, dorsal view; c, nasale and subnasale, ventral view; e, urogomphi, dorsal view; g, left urogomphus, lateral view. 0b, h, 1, C. riparius (Fabricius) : b, nasale and subnasale, ventral view; h, left urogomphus, lateral view; 7, ninth abdominal segment, dorsal view. d, f, C. funebris Candeze: d, nasale and subnasale, ventral view; f, left mediotergite of fourth abdominal segment, dorsolateral view (showing complete complement of setae; the num- bered setae are found in all known Cryptohypnus larvae). NOw et LARVAE OF THE ELATERID BEETLES—GLEN 245 --ret —-ret g pereqyeyly a TT F My, Fic. 39.—Hypnoidus dubius and Hypnoidus musculus. a-e, h, i, Hypnoidus dubius (Horn): a, frontoclypeal area, dorsal view; b, gular area, ventral view; c, left mandible,, dorsal view; d, third abdominal segment, ventral view; e, left mesothoracic leg, mediocephalic view; /, ninth and tenth abdominal segments, lateral view; 7, ninth abdominal segment, dorsal view. f, g, H. musculus (Eschscholtz): f, urogomphi, dorsal view; g, left mandible, dorsal view. 246 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 40.—Eanus decoratus and Melanactes densus. a-e, Eanus decoratus (Mannerheim) : a, right mandible, dorsal view; b, third abdominal segment, lateral view; c, third abdominal segment, ventral view; d, ninth and tenth abdominal segments, lateral view ; e, ninth abdominal segment, dorsal view. f, Melanactes densus LeConte: larva, dorsal view. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 12 NOTE ON ROWE S sEEC TROSCOPIC METHOD FOR tae DeTERMINA TION OF AQUEOUS Vac ORIN THE ATMOSPHERE (WitTH ONE PratTe) BY LB: ALDRICH Director, Astrophysical Observatory (PUBLICATION 3989) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION SEPTEMBER 20, 1949 The Lord Baltimore Press BALTIMORE, MD., U, 8. A. NORD ON BOWLES SPECTROSCOPIC. METHOD FOR THE DETERMINATION OF AQUEOUS VAPOR IN THE ATMOSPHERE By L..B. ALDRICH Director, Astrophysical Observatory (WiTH ONE PLATE) In 1912 F. E. Fowle* developed a spectroscopic method for deter- mining the amount of water vapor in a column of atmosphere verti- cally above the observer. This quantity he called the “precipitable water,” and he defined it as “the depth of liquid water which if in the form of vapor would be contained in a column of air of the same cross-section reaching vertically to the limits of the atmosphere.” Fowle’s method uses three infrared water-vapor bands, viz, p(A= 9354), $(A=1.134) and wW'(A=1.47p). With spectrobolometric energy curves obtained at Washington by passing the radiation from a bank of Nernst glowers through a known quantity of water vapor, he studied the depths of these absorption bands as affected by changes in width of bolometer strip, slit, and other variables, The depths as : L deflection at bottom of band expressed in the ratio deflection at smooth curve above the band (or in abbreviated form, p/psc, 6/dsc, and W’/W'se) he correlated with known quantities of water vapor in the path of the beam. The amount of water vapor, expressed in centimeters of precipitable water, was determined from the length of path and from wet- and dry-bulb ' readings taken at many points along the path of the beam. The accuracy of these water-vapor determinations he checked by absorbing with phosphorus pentoxide and calcium chloride the water vapor in known volumes of air along the path. Thus he developed a table giving for standard conditions of definition and purity of the spec- trum, as controlled by slit width, bolometer strip width, time of swing of galvanometer, etc., the precipitable water corresponding to given values of p/pse and $/d¢sc. This table has since been extensively used in the solar-constant work of the Astrophysical Observatory, particularly in daily determinations of the short-method function F. (see Ann. Astrophys. Obs., vol. 6, p. 66). 1 Astrophys. Journ., vol. 35, p. 149, 1912; vol. 37, p. 359, 1913. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 12 bo SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS rn In 1947 the Astrophysical Observatory temporarily established at Miami, Fla., a solar field station, to carry out certain radiation mea- surements under contract with the Office of the Quartermaster \2 ‘eh i SC (PREC. WATER x/0) IN CMS. PRECIPITABLE WATER RATIO BOTTOM OF BAND SMOOTH CURVE Fic. 1.—Fowle’s precipitable-water curves. (See table 1.) General. These measurements are a part of studies of the causes of deterioration of textiles, and at Miami are conducted in cooperation with the South Florida Test Service and with the General Motors Corporation on whose test field the observations are made. A part NOD LZ DETERMINATION OF AQUEOUS VAPOR—ALDRICH 3 of the equipment for this work was brought from Burro Mountain near Tyrone, N. Mex., where for some years the Smithsonian Institution maintained a high-altitude solar station. The circumstances of a Smithsonian solar field station operating at a wet, sea-level location is unique. The only previous spectrobolo- metric set-up at sea level was the original Washington equipment of some 40 years ago, which Fowle used in his precipitable-water studies above mentioned. The Miami set-up thus offered opportunity to check the correctness of Fowle’s precipitable-water curve. \ \ N ‘ N A va N te \ x \ \\ \\ HY, Mi Fic. 2.—Path of beam. L, light source; Ms, M2, concave mirrors, I-m. focus; C, coelostat plane mirror; S, slit; B, bolometer. We originally planned to do this in June 1948, at which time of year there is normally a maximum of atmospheric humidity. Unavoid- able delays occurred, however, and bolographs were not obtained until January 1949. The following summarizes the work: A 500-watt projection lamp was placed in the focus of a 9-inch- diameter aluminized mirror of 1 meter focal length. The assembly was mounted on a movable table in the field adjoining the station building. The approximately parallel beam from the mirror fell upon the second mirror of the coelostat and from there passed through the spectrobolometer, housed in the station building (fig. 2). In Smithsonian solar-constant observations, standard conditions of slit width, bolometer strip width, and time of swing of the gal- 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III vanometer suspension are maintained. Effort is made to keep the definition constant, as measured by the depth of the Fraunhofer line “h” (A=.4102n). The ratio of bottom of this band to the smooth curve over the top is kept at .845. In the bolographs of the present work these standard conditions were fulfilled. Constancy of the lamp source was assured by the interpolation of an electronic voltage regulator capable of holding the potential across the lamp uniform within 1/10 percent. Bolographs were made on January 28 and 29 for distances lamp source to bolometer strip vary- ing from 19 to 80 meters. Repeated wet- and dry-bulb readings were taken at various positions along the path of the beam both indoors and outdoors, during the time bolographs were being recorded. Hand- aspirated Friez psychrometers were used, checked periodically against a sling-type psychrometer. Figure 1 gives Fowle’s curves for p/psc and $/¢sc. The dotted lines extend the curves to zero water vapor. The individual points of January 28 and 29 lie on the extended curves within experimental error. Actual deviations from Fowle’s values are given in table 2. As would be expected, the deviations are generally larger as the deflections become smaller. It is gratifying that Fowle’s curve appears substantially correct for the water-vapor range here tested. On January 28 the average air temperature was 73° F. and relative humidity 80 percent. On January 29 air temperature was 67° and relative humidity 9o percent. 6V6l “62 AYVNNVE AO SHA VYDOIOgG adic h CONS AEE Rea ON SNOILOS1109 SNOANVI1S0OSIN NVINOSHLIWS NO. 12 DETERMINATION OF AQUEOUS VAPOR—ALDRICH 5 TaBLE 1.—Fowle’s standard table. Precipitable water vs. depth of band (For standard conditions of solar-constant observations: slit width, .3 mm.; time of single swing of galvanometer, 1.4 sec.; ratio of bottom of the Fraun- hofer “th” line (A =.4102 ») to smooth curve = .845.) Precipitable Precipitable water water p/pse b/dse in cm, p/pse d/se in cm. .210 .104 6.34 .510 381 1.60 .220 sti 6.09 .520 .392 1.51 .230 .119 5.85 .530 .403 1.42 .240 3127, 5.62 .540 414 1.34 .250 .136 5.40 .550 425 1.26 .260 145 5.19 .560 436 1.19 .270 154 4.99 570 .448 1.12 .280 .163 4.80 .580 .450 1.05 .290 172 4.61 5900 .470 .98 .300 181 4.43 .600 481 .Q2 310 .190 4.25 610 493 87 .320 .200 4.08 .620 504 82 -330 .209 3.91 .630 515 Ti -340 .217 3.75 .640 .526 72 350 .226 3.59 .650 537 .67 360 234 3.43 .660 548 63 .370 242 3.28 .670 .560 59 .380 251 3.13 .680 ASA 55 .390 259 | 2.08 .690 583 51 .400 207 2.84 -700 504 Az .410 .276 2.70 710 .606 43 .420 285 PSG) .720 618 .40 .430 294 2.44 .730 .630 .37 .440 304 2:22 .740 642 34 .450 315 2.20 .750 655 31 .460 326 2.09 -700 .667 28 470 337 1.98 770 .680 25 .480 347 1.88 .780 .693 2% -490 -359 1.78 .500 .370 1.69 Date 1949 Jan. 28 Jan. 29 SMITHSONIAN MISCELLANEOUS COLLECTIONS TABLE 2—Summary of January 28 and 29 data Total distance source to bolometer (meters) 10.44 30.51 51.81 79.66 19.44 30.51 51.81 79.66 Maxi- mum deflec- tion (cm.) 10.80 8.28 3-13 1.36 II.10 2.92 1.16 Bolo- graph number I II Ill IV Vit Band Gomer ash Gs) Reh as) > Gv Ss ~ = Ss =) S55): SOo5 Ss co 6 > S: > 6: > “6: > Ratio -939 .889 .936 885 -905 849 .QI2 848 876 ind 77 .867 -790 816 .740 825 .740 .949 .902 -947 .899 916 851 .O16 854 877 794 884 .802 823 742 .839 .756 Precipi- table Fowle’s curves (cm.) .032 | .032 f 034 | .033 J .052 | .050 f .048 | .050 f .080 | .096 f .086 | 082 f 145 | 144 f -132 | 144 J .027 nN .026 f .0275 | 020) .046 | .046 f 045 | .045 J .073 | .076 f 074 | 074 J 135 | 133 J ALS ity) Total Ob- served value (cm.) .032 045 £045 073 075 a128 .11Q VOL. IIT Difference oar al ) ©) a i=) ++ + oe ce 8 w | as ° ia) + .o12 + .o10 — .006 — .002 f +.043 Differences | — .046 a ee SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 13 Roebling Fund SHORT PERIODIC SOLAR VARIATIONS ND Shik bE MPEP RATURES OF WASHINGTON AND NEW YORK BY €, G. ABBOT Research Associate, Smithsonian Institution (PusiicaTion 3990) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION OCTOBER 4, 1949 The Lord Baltimore Press BALTIMORE, MD., U. & Je Roebling Fund SHORT PERIODIC SOLAR VARIATIONS AND THE TEMPERATURES OF WASHINGTON AND NEW YORK By C. G. ABBOT Research Associate, Smithsonian Institution In my paper read before the National Academy of Sciences in April 1949, fairly successful predictions of 55 minima of Washington temperatures for the year 1948 were discussed. The 55 dates in question were the dates when the period of 6.6456 days * would recur in the same phase as on January 17.0000, 1946. It was privately indi- cated to me, after the delivery of my paper, that certain correlation studies which had been made on New York City temperature de- partures raised doubts if similar results would have been obtained for that station. By the kindness of E. J. Christie, Meteorologist in Charge at New York, Weather Bureau forms 1030 giving departures from normal temperature for every day from January 1928 to date were furnished for me to study this question. WHY EXPECT SUCH A REACTION? Before proceeding further, let us refresh our minds on the reasons for supposing that there should be a period of 6.64561 days in meteorology.” Solar radiation is the source which maintains the earth’s temperature and other meteorological phenomena. If the solar radiation is variable, these must be expected to vary. Daily observa- tions by Smithsonian observers carried on at Montezuma, a mountain 9,000 feet high in the Atacama Desert of northern Chile, show that small fluctuations of the sun’s output of radiation do occur. A statistical study of them showed that there is an approximately regu- lar solar variation of about 6% days’ period. The departures from normal temperatures at Washington between the years 1910 and 1945 were tabulated in 63-day intervals for the months of May and November, separately. On plotting the 63-day 1 As shown below a small correction now alters this period to 6.6485 days. 2 See Smithsonian Misc. Coll., vol. 107, No. 4, Apr. 4, 1947. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 13 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS ier curves for May, it was seen that a decided indication of a variation of about that period occurred, but sometimes from 1 to (rarely) 3 days before or after it was expected. However, upon scanning the whole 45 years of May values, it was clear that the period was either a little shorter than 63 days (best value found to be 6.6456 2 days which is about 1/55 of a year) or else it must be about 1/54 or 1/56 of a year. Now turning to the tabulation for November, it was found that only the period 6.6456* days could serve. For if either 1/54 or 1/56 of a year were adopted, the November curves would be a half period out of phase with those of May. Afterward, the solar-constant values of Montezuma, 1924 to 1945, were critically studied. They proved to exhibit this period of varia- tion, and with perfect regularity. The displacements of from 1 to (rarely) 3 days in phase in Washington temperature departures from the normal do not occur in the solar variation. Considering the complexity of the earth, its atmosphere, and meteorological reac- tions, it is not surprising that such displacements of the phases of a terrestrial response to regular solar pulses should occur. But they impair the value of the solar period for forecasting purposes. It was shown, however, in Smithsonian Miscellaneous Collections, vol. 107, No. 4, that the temperature departures caused by this periodic solar variation range from 2° to 20° F. at Washington and also at St. Louis and at Helena, Mont. PROCEDURE OF INVESTIGATION Proceeding with the investigation of New York City temperature departures, I made monthly tables of the four or five recurrences of the 6.6456-day period in each month. These covered 21 years from 1928 to 1948, making 1,154 separate recurrences in all. The tables were based on January 17.000, 1946, so as to be comparable directly with the Washington work reported in Smithsonian Miscellaneous Collections, vol. 111, No. 6. The following sample, table 1, is for May 1929. It will be noticed that only six values occur in lines 1 and 4 of the table and that the vacancies are distributed to the last and first columns. These adjustments take into consideration whether the fractions in the phase dates exceed 0.5 or not. The vacancies are distributed so that, in the year, as many fall in the first column as in the last. Proceeding in this way the New York data were all tabulated from 1928 to 1948 by individual monthly means. It was NOS LS SHORT PERIODIC SOLAR VARIATIONS—ABBOT 3 then noticed that maxima and minima in the seven-column means occurred progressively later as the years went on. CORRECTION OF THE PERIOD By graphic methods applied to both maxima and minima, and to several different months of tabulation, it appeared that the secular displacement of features just referred to amounted to 3 days in 18 TaBLe 1.—New York City temperature departures, May 1929, arranged in the 6.6456-day period April May June Phase dates .... 28.6936 5.3392 11.9848 18.6304 25.2760 0.9216 Temperature departures .. — 3° — 3° its 1° —2° —6° 5 I —2 —8 —-6 —5 —3° 5 Qe 10 2 =7 0 2 —10 —I4 —4 —9 —I1 —I 2 4 10 12 10 15 Mean.. TS 1.0 — 2.2 — 0.2 0.4 — 3.4 27, years. As there are 55 recurrences of the period each year, this dis- 3 placement corresponds to Agee = 0.0030 days per cycle. It is more 3) convenient to use numbers ending in 5 than in 6, and the accuracy of the determination does not justify the inconvenience, so, instead of 0.0030, a correction of 0.0029 was applied, making the corrected period 6.6485 days. By subtracting and adding this number of days many times to January 17.0000, 1946, a tabulation was prepared giving dates of all expected minima of temperatures at Washington from 1928 to 1948. Applying this table also at New York it would be found whether minima, occurring there by the effect of the periodic solar variation, fall on the same dates as at Washington. PREPARATION OF THE DATA Using Weather Bureau forms 1030, the departures from normal temperature were tabulated, exactly as in table 1, for both Wash- ington and New York, from January 1928 to December 1948.: In these tabulations the newly corrected period 6.6485 days is used. As stated above, the tabulation is based on January 17.0000, 1946. As the reader will note in table 1, this arrangement results in tables of seven columns, and these were labeled 1 to 7. Each month— 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT January for instance—yields first a table of either four or five lines, depending on whether some January days were required to complete lines of December or February, or whether some days from one or both of those months were required to complete lines of January. Mean values having been taken, as in table 1, there resulted a new table of 21 lines, of 7 columns each, for each one of the 12 months of the year, the entire 12 tables covering the interval 1928 to 1948. FREQUENCY DISTRIBUTION From the 12 tables of mean monthly temperature departures were now read off from their 21 lines the days (1 to 7) on which maxima and minima of temperature occurred. Where identical mean depar- tures came on more than one day, the reading was appropriately split up, as, for instance the same minimum occurring on days I and 7 was tabulated as 4 at I and 3 at 7. Terrestrial responses to solar impulses lag behind their solar causes. Thus, for example, the coolest and warmest parts of the day occur several hours after midnight and noon, respectively. The lag differs from place to place, and from time to time, depending on terrestrial complexities. So it should not be expected that maxima and minima of temperatures at New York and Washington, due to a periodic solar change, would necessarily be coincident, or that at either sta- tion they would always be found at the same columns of the tables. If, as hitherto generally supposed by meteorologists, weather is almost wholly governed by terrestrial influences, apparently acci- dental as to timing, then a study of the frequency of maxima and minima in my tables should show no marked preference for any columns over the others. But if, as indeed will be shown below, the regular solar periodic pulse is a principal cause of weather, then, despite the interference by terrestrial complexities, there should be found marked preferences for certain columns as seats of maxima and minima. Chance is largely subordinated by multiplicity in this study, for each month of the 12 monthly tables results from 5 : 20 22 =g6 recurrences of the period. It was presumed that the lag of temperature response might differ in different parts of the year. Hence the 12 months were tabulated separately. Moreover, a great collection of unpublished studies on the terrestrial responses to 14 long-period solar variations has shown that such responses are apt to differ greatly with sunspot frequency. Hence, in each of the 12 monthly tables, the years 1928-30, 1936-41, NO. 13 SHORT PERIODIC SOLAR VARIATIONS—ABBOT 5 and 1946-48 were kept separate from the years 1931-35 and 1942-45. In anticipation, it may be remarked that slight displacements of maxima and minima seem to attend both the time of the year and the epoch of the sunspot cycle. But these displacements are so small, and the effect of the 6.6485-day period is so pronounced, that a direct mean of all the evidence, for all months and all years, shows dis- tinctly the main features of the frequency of distribution of the temperature departure features. TABLE 2.—Frequency of temperature minima in the seven columns No. of Lines Station I 2 3 4 5 6 7 years TNE. crave Wew York 952 “30° 30 42 #25 26 47 21 Peseta ree Bae Washington 44 35 23 46 26 23 = &«55 21 Gide Bhat e rc Ney More As mh Ao) AS V/s nO 12 Tse Ae’ eee 3 Washington 23 18 15 31 WA jie AY 12 eee Leeatceet Ae New York 26 15 10 17 8 13 21 9 (Car aierc een Stes Bitar Washington 21 17 Saas 14 5.) 6 28 9 7 Ae A Ste, Foe INe@wa Vonks 2/33) 1-259) 1467, ©2108) 1-42) 1.08) 2:17 I Sigs. SOTERA Geer Washington I.92 1.50 1.25 2.58 1.00 1.50 2.25 I GB Om oo ioe New York 2.67 1.67 1.11 1.89 089 1.44 2.33 I TOM eros Washington 2.33 1.89 0.89 1.67 1.56 0.56 3.10 I NER eg eee Secs wat ee Gee ep ee EXxPLANATION.—Lines 1, 2 cover monthly means for all years, 1928-1948. Lines 3, 4 cover monthly means for years of sunspot maximum. Lines 5, 6 cover monthly means for years of sunspot minimum. Lines 7-10 cover as lines 3-6, but are general averages for 1 year only. Line 11 shows when years of sunspot minimum have higher fre- quencies. In table 2 and figure 1 are given the frequencies with which minima in all the monthly mean tables combined (similar to table 1) fall in the seven different columns of the tabulations. For curiosity’s sake the table also gives the average numbers per month of occur- rences of mean minima for sunspot maximum and sunspot minimum years separately. Reducing the values to equal numbers of years, it will be seen from these latter tabulations that for both Washington and New York there is a tendency in sunspot minimum years for minima to occur with greater frequency in columns 1, 2, and 7, and lesser frequency in columns 3, 4, 5 and 6. This shows that the prin- cipal solar period has better control over terrestrial disturbances when sunspots are at minimum. But ignoring this subordinate result, the main results of the tabulation show that: 1. There is a great and nearly equal preponderance of frequency for both stations in columns I and 7, so that the minimum falls at about half a day after 7 in both cities. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS set There is a strong frequency in column 4, showing that another 6.0485 regular periodic solar variation exists of = 3.3242 days in period. Its phases coincide with those of the primary period when- ever possible. 3. There is no appreciable difference in lag between Washington and New York in response to the solar variations. In short, the two sta- tions behave nearly alike in all respects. WASHINGTON VAG MAGNITUDE OF THE EFFECT TRG 10s Owing to terrestrial interferences in lag and otherwise, and to the interference caused by the existence of the scondary solar period of half the length of the primary one, the positions of maxima and minima fluctuate. Comparatively seldom does a whole month go by without shifts of 1, 2, or rarely 3 days in the place of minima. On this account the mean monthly values seldom show the full measure of the effect of the solar change on terrestrial temperatures. How- ever three months have been selected from many among the tempera- NOLONIHS VM MYOA MAN ABBOT VARIATIONS | ! sete tala xs —_—_{—__—- 02 SHORT PERIODIC SOLAR 13 | | | NO 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Tal ture departures for each of the two stations, when the regularity of the periods was little disturbed during the whole month. Months were chosen fairly well distributed throughout the year. Plots of these temperature departures are given in figure 2. By drawing straight lines across the bottom of each hump, values have been read off giving roughly the numbers of degrees by which the temperatures were raised by the solar influence. As is well known, the tempera- ture changes of all sorts are much less in July than in the cooler months of the year, in the Eastern United States. Hence it is not surprising that this appears in figure 2 and in table 3. TABLE 3.—Average magnitude of the temperature fluctuations in degrees F. New York Washington Jan! 1032, 12°26 Feb. 1928, 15°2 Mar. 1936, 14°2 Mar. 1946, 14°0 July 1938, 7°8 Nov. 1943, 12°4 By inspection of these exhibits one sees that throughout most of the year, in Washington and in New York, the temperature is af- fected either four or five times each month by a change in solar radiation, by amounts ranging from 10° to 20° F. CONCLUSION Contrary to the suggestion referred to at the beginning of this paper, Washington and New York respond almost alike to the short regular variations of solar radiation. Two such variations are known. The primary one has a period of 6.6485 days, and the subordinate one half that length. Thus the primary solar change recurs either four or five times each month. Temperature changes of 10° to 20° F. attend its every recurrence during most months of the year, but dur- ing summer the effect is somewhat less. The phases of the tempera- ture changes appear to be the same at Washington and New York. —e } i wis | r i i } . i i f i . ~ N : é ; j t a? . | a : Ps ‘ 4 i} *t t ‘ t s— 7 — . rr \ 9 * \ it e' = * : ' 5 > a 7 7 ‘ P| ‘ ‘ “4 4 * ? i ME Beryl @ on _ “Sy 7 : -~ Ay : > « . : ) : , . ; Ls x " . 2 1 vy rom erin © 24 y Ae 7 1 m4 } 7 ® é adn 8 et wes pty 7 “wo 1 ee am i. - 5 i ae ad A oiiaciet iis Se ae Pig aah ne OP aa) rig aa: re + SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 14 Hodvgkins Fund Mine eo ft StL VER- DISK odie IOME TER (WitH ONE PLATE) BY L. B. ALDRICH Director, Astrophysical Observatory (PusiicaTion 3991) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION DECEMBER 8, 1949 “an ; wa eT) ON Cn as Se me The Lord Baltimore Dress a BALTIMORE, MD., U. S. A. e ays i} ( | ve : | oa Hodsgkins Fund Dr ABBOT SILVER-DISK PYRHELIOMETER By L. B. ALDRICH Director, Astrophysical Observatory (With ONE PLATE) DESCRIPTION OF INSTRUMENT In 1922 Dr. C. G. Abbot published a paper (The Silver-Disk Pyrhe- liometer, Smithsonian Misc. Coll., vol. 56, No. 19) describing his pyrheliometer, the method of use and possible errors, and listing the constants of the various instruments. Since 1922 several modifications of the instrument and of its method of use have been adopted, and some, 50 additional pyrheliometers have been prepared and sold to interested institutions throughout the world. It therefore seems ad- visable to bring Dr. Abbot’s 1922 discussion up to date. Volume 2 of the Annals of the Astrophysical Observatory of the Smithsonian Institution (p. 36) describes the original mercury pyr- heliometer which Dr. Abbot built in 1902 and from which the silver- disk pyrheliometer developed. In 1909 the form of instrument shown in cross section in figure 1 was adopted. The following description of it is taken from Dr. Abbot’s paper (Smithsonian Misc. Coll., vol. 56, Nok) : The silver disk, a, shown in cross-section [fig. 1], is bored radially with a hole to admit the cylindrical bulb of a thermometer, b. The hole in the disk has a thin lining of steel, so that a small quantity of mercury may be introduced without alloying the silver, in order to make a good heat conduction between the silver disk and the thermometer bulb. A soft cord soaked in shellac is forced down at the mouth of the hole to prevent escape of mercury, and a ring of Chatterton! wax is sealed over the outside of the cord to make the closure more perfect. The thermometer, b, is bent at a right angle, as shown, in order to make the instrument more compact and less fragile. A nickeled brass tube (shown partly cut away in the figure) supports and protects the thermometer. A slot is cut in the right-hand side of the support tube throughout almost its whole length to permit the reading of the thermometer. At the top of the support tube a short piece is removable, in order that heat may be applied to the over- flow bulb of the thermometer, to dislodge mercury which sometimes collects there 1 Picein wax has been used in recent years. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 14 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III a / JOSS SSSA ot tore 57 Fic. 1.—Abbot silver-disk pyrheliometer. NO. 14 ABBOT SILVER-DISK PYRHELIOMETER—ALDRICH 3 during transportation. The thermometer is graduated to tenths of degrees centigrade from —15° C. to +50° C. Two points, 0° and +50°, are first marked on the stem by the makers, and then the thermometer is graduated by equal linear intervals without regard to the variations of cross-section of bore of the stem. Before insertion in the instrument, a careful calibration of the thermometer stem is made. The silver disk, a, is enclosed by a copper cylindrical box, c, halved together for convenience in construction. Three small steel wires, not shown in the figure, support the silver disk. These wires lie in the plane of the center of the disk at 120° intervals apart. Midway between them are three brass screws, not shown, which may be screwed through the walls of the box, c, up to their heads. These screws in that position clamp the silver disk tightly. Their purpose is to prevent the breakage of the thermometer if jarred during trans- portation. These screws must be loosened during observations. The copper box, c, is enclosed by a wooden box, d, to protect the instru- ment from temperature changes. This box is also halved together and fas- tened by long wood-screws, one of which is seen near the letter, d, in the figure. Sunlight may be admitted through the tube, e. This tube is provided with a number of diaphragms, f:, fz, fs, having circular apertures. The aperture, fs, nearest the silver disk is slightly smaller than the others, and slightly smaller than the disk itself. Thus it limits the cross-section of the sunbeam whose intensity is to be measured. The entire interior of the tube, e, the box, c, and the silver disk, a, are painted dead black with lamp-black mixed in alcohol, with a little shellac added to cause the lamp-black to stick. To secure a fine, even coat, the mixture is filtered through cheesecloth before applying, and on the disk it is warmed with an alcohol lamp until the brush marks disappear. A rotatable shutter, g, with three nickeled parallel metal plates, h h h, is pro- vided for cutting off the sunlight as desired. The top of the tube, e, carries a screen, k, large enough to shade the wooden box, d. This screen also sup- ports the thermometer tube, and the axis of the exposing shutter just men- tioned. A small hole in the part which supports the thermometer admits a little guiding beam of sunlight, i, whose use is to assure the observer that the instrument points toward the sun. The pyrheliometer is carried upon an equatorial stand, as shown in the figure. A worm and wheel mechanism is provided for following the sun. No clock- work is needed, as it is sufficient for the observer to move the worm slightly two or three times a minute. In 1927 two alterations were adopted: (1) the tube e was length- ened and (2) the base was enlarged to counteract top-heaviness due to the longer tube. These changes are shown in the photograph, plate I. The short tube form of 1909 when the shutter was opened exposed each point on the silver disk to a cone of sky 10°38’ in diameter. Since the sun subtends a diameter less than 1/20th as great, the sky area exposed is at least 400 times that occupied by the sun. In very clear skies the radiation from this area of sky is negligible as compared to that from the sun. But measurement at Mount Harqua Hala in 1925 ? 2 Annals of the Astrophysical Observatory of the Smithsonian Institution, vol. 5, p. 83. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Ii! in a very hazy sky indicated that the readings were as much as 23 percent too high. To diminish the error from this source all silver- disk pyrheliometers made after 1926 carried tubes 32 cm. long in place of the original 15 cm. The exposed sky area was thus reduced from .0043 hemisphere to .oo13. As opportunity has arisen the fol- lowing older instruments have been similarly modernized by substi- tuting the longer tube and enlarging the base: S.1. Nos. 1, 5, 16, 17, 26, 31, 41, 42, and 44. In hazy skies the error due to sky radiation now seldom exceeds 4 percent. While it is possible, from a study of the dimensions and physical properties of the instrument, to reduce its readings to heat units, there are uncertainties in thus using it as an absolute standard. We have preferred to consider it as a secondary instrument, the constant of each individual pyrheliometer being determined by careful com- parisons with Smithsonian standard instruments.* Readings of the silver-disk instrument are proportional to the intensity of radiation of the sun, and are comparable one with another at all times and places. Multiplying the corrected readings by the constant of the in- strument reduces them to true heat units. DIRECTIONS FOR USE These directions, with minor alterations to conform with our pres- ent practice, are taken from Dr. Abbot’s paper, mentioned above. SETTING UP 1. If the mercury column of the thermometer is broken, remove the little screw at the side near the upper end of the nickel-plated tube, and take off the upper portion of the tube. Then heat the exposed stem cavity of the thermometer gently in a smoky flame (a match flame is good) until the mercury is expelled from the cavity. Then, holding the thermometer vertical, shake the instrument repeatedly with a downward jerk until the mercury columns join. 2. Remove the two little ivory or fiber plugs (using pliers if nec- essary) and unscrew the two brass screws under them, and also un- screw the third similar screw seen through the trunnion on the other side of the case. About three complete turns of each screw is proper. Insert the two ivory plugs. When packing the instrument again for 3 Descriptions of Smithsonian absolute standard pyrheliometers are given in Annals of the Astrophysical Observatory of the Smithsonian Institution, vol. 3, p. 52; vol. 6, p. 5; Smithsonian Misc. Coll., vol. 87, No. 15; vol. 92, No. 13; vol. 110, No.5; NO. 14 ABBOT SILVER-DISK PYRHELIOMETER—ALDRICH 5 a journey, screw in the three screws as far as they will go. Their pur- pose is to clamp the silver disk to protect the thermometer during transportation. 3. Unscrew the two pivots from the sides of the ring of the mount- ing, insert the pyrheliometer so that the thermometer is not next to the worm wheel, and screw in the pivots. 4. Unclamp the half ring and set the polar axis approximately for the latitude of the place. The thermometer should be next to the upper end of the axis. ADJUSTMENTS 5. For quick adjustment in right ascension guide the pyrheliom- eter with one hand and loosen the lower right-hand milled screw (as seen from the upper end of the polar axis). The worm may then be lowered out of engagement with the wheel and the change made. 6. To follow the sun, adjust in right ascension and declination until the sun shining through the little hole in the upper plate forms its image on the scratched spot on the nickeled piece below. When ex- posing to solar radiation rotate the worm screw a little (about once every half minute) to follow the sun. 7. When about to observe, push aside the cover, leaving only the shutter to shade the silver disk. When through with each series of readings, close the cover to keep out dust. OBSERVATIONS 8. When reading the thermometer the observer should hold his head so that the reflection of each dark line of the scale near the degree to be observed, as seen in the mercury thread, is coincident with the corresponding dark line. This prevents parallactic errors of reading. 9g. Having adjusted the instrument to point at the sun and opened the cover, read the thermometer exactly at 20 seconds after the be- ginning of the first minute. Read again after 100 seconds, or at the beginning of the third minute, and immediately after reading open the shutter to expose to the sun. Note that the instrument is then cor- rectly pointed. After 20 seconds read again. After 100 seconds more (during which the pointing is corrected frequently), or at the begin- ning of the fifth minute read again, and immediately close the shutter. After 20 seconds read again. After 100 seconds read again, or at the beginning of the seventh minute. Continue the readings in the above order, as long as desired. Readings should be made within $ second of the prescribed time. Hold the watch directly opposite the degree to be observed, and close to the thermometer. Read the hon dredths of degrees first, the degree itself afterward. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. / EI 10. For results not requiring the highest accuracy the above de- scribed method of reading is satisfactory. As the result of long ex- perience, we now use for greater accuracy several refinements of this method, as follows: (a) A special reading glass is used. It consists of a small eyepiece of about 4 cm. focal length, mounted so that it can easily be held against and moved along the thermometer stem. In the focus exactly in the center of the field is a sharp needle point. By taking readings when the needle point is opposite the top of the mercury column, parallax errors are eliminated. (b) Any simple device to beat regular intervals (such as I, 5, or 10 seconds) permits the observer to concentrate on reading the ther- mometer instead of trying to read both watch and thermometer at the same time. Such a device also eliminates possible error due to eccen- tricity of the second hand of the watch. (c) The instrument is set out in the sun at least 15 minutes be- fore starting to read, and the shutter opened to the sun for about I minute during this period. In making a series of observations, the second set of readings is started 20 seconds after completing the first set of six readings. Thus a 4-minute shaded period occurs be- tween each 2 minutes of exposure. Each set of six readings is quite independent. 11. For example: Reading .... I 2 3 4 5 6 me) co eke 11°55™20° 57™00° 57720" 59™00" 59™20" 001700" Reading, 524.2) 1512 14°25 14°80 17°58 17°36 16°09 Condition .. Shaded Exposed Shaded Air temperature 15°. Pyrheliometer “S.I.Q.” 12. Subtract readings (2) from (1) ; (3) from (4) ; (6) from (5). 13. Take the algebraic means (NZ) s(5) 2G) 2 and to them add [(4)—(3) ]. 14. Call this result R,. Find roughly the mean temperature T, dur- ing the interval of exposure (3) to (4). 15. AddtoR, the percentage correction for graduation furnished with the instrument, then, after correcting, add to R,, K[ (Ti—30°)R, |]. K is a constant furnished with the instrument. If the prevailing tem- perature of the air differs much from 20°, add 0.0014 R for each 10° the air temperature falls below 20°. The result (which we will call R*,) is the final rate of rise per 100 seconds during the exposure (3) to (4) as reduced to the standard bulb temperature of 30°, and standard stem temperature of 20°. NO. I4 ABBOT SILVER-DISK PYRHELIOMETER—ALDRICH 7 16. (Note.) The approximate method of procedure stated in (13) and (15) is much easier than the exact method, and having been found by experiments to yield closely comparable results under all circumstances of use, within the error of measurement, it has been adopted, and the standardization of the instrument is made by this method. 17. To reduce the result R*, to standard calories per square centi- meter per minute, or to the Smithsonian scale of 1913, multiply by the factors furnished from the Smithsonian Institution with each in- strument. 18. Example of reduction: INSEL, ays aieie scones I 2 3 4 5 6 InGEYbIS GoooeeBO ae 15°12 14°25 14°80 17°58 17°36 16°09 Ditterences: ........- 0°87 2°780 1°27 Cooling correction. . Sages 1°070 18 Recs pee oe ae 3°850 “LR Ge SRS eee aisteye 16°2 Scale correction ... —0.0020R:= —o0°008 KR(T—30°) ...... — (0.0011) (13.8)R:i= —0o0?%058 Air correction ..... --0.0007R:= +0°003 RSs ltyebotetsicyeiahers avalisiets i, ele 3°787 PRECAUTIONS Constant watchfulness and care are needed to keep the instrument in best condition. For example, it should not be inverted or jarred, thus keeping the mercury column and the wax bond between thermometer and silver disk intact. The blackening of the silver disk should be examined periodically. If there is evidence of specks or spottedness, the surface should be carefully brushed with a soft camel’s hair brush fastened to a long slender handle. The cover at the end of the tube should be kept closed when not in use. With care the blackness remains unchanged over many years. Our two substandard silver-disk instruments, A.P.O. No. 8pis and S.I. No. 5 are evidence of this. Repeated comparisons against the absolute water- flow pyrheliometer over a period of 20 years have shown no evidence of change. A silver-disk pyrheliometer ready for shipment, including stand- ardization and boxing, costs the Smithsonian Institution about $200. While it is not desired to manufacture them extensively, it has been our practice to prepare and sell these pyrheliometers to individuals and institutions likely to use them for valuable and regular solar ob- servations. The special reading glass above mentioned can be fur- nished for $10 additional. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Adopted constants of silver-disk pyrheliometers Smithsonian pyrheliometry, scale of 1913 Instru- Present ment constant Sh be peers 0.3733 Sele eee 0.3743 Sola iyiekice 0.3625 SHG termes 0.3713 Sli Genapede 0.3715 SHR Oise: oan 0.3666 SU, Ata 0.3638 Ssh58 tarsi: 0.3774 Seo naan 0.3738 SHO: 0.3762 SH Blea oead 0.3769 SHIRM IS 65 ee 0.3631 Sa Ia i ee tee 0.3617 Sele Vases 3 0.3721 Siley UG isars, sues 0.3609 Sak, T6=ac:. 0.3634 Sale 7 sacs 0.3629 SA Shee. 0.3774 Sule uO sae doG 0.3737 SA 20 ye etstare 0.3057 Sul 2tienccien 0.3711 Saeraon aes 0.3778 Siew 2S even 0.3683 Sule 24 stat 0.3713 Sle zo aoe 0.3717 Location and remarks U. S. Weather Bureau, 1910. Modernized, 1929. Observatorie Physique Central Nicolas, St. Peters- burg, Russia, 1910 (loan); National Observatory, Rio de Janeiro, Brazil, 1912. Conservatoire des Arts et Métiers, Paris, France, 1911. Royal University of Naples, Italy, 1911. Destroyed in World War II. U. S. Department of Agriculture, Physical Laboratory, 1910. Modernized, 1931. On loan to Astrophysical Observatory since 1926. Oficina Meteorologica, Buenos Aires, Argentina, I9I1. Do. Central Observatory, Madrid, Spain, 1912. Imperial College of Science and Technology, London, England, 1911. Koniglich Preussisches Meteorologisches Institut, Ber- lin, Germany, IQII. Meteorological Observatory, Teneriffe, Canary Is- lands, IQII. Koniglich Preussisches Meteorologisches Institut, Ber- lin, Germany, IgI2. Schweizerische Meteorologische Centralanstalt, Zu- rich, Switzerland, 1912. University of Toronto, Toronto, Canada, 1912. Mod- ernized, 1937. National Bureau of Standards, Washington, D. C., 1912. University of Arizona, Tucson, Ariz., 1912. Modern- ized, 1929. Sent to Service Botanique, Tunis, 1932. Harvard College Observatory Station, Arequipa, Peru, 1912. Sent to Smithsonian Observing Station, Calama, Chile, 1919. Modernized, 1920. Observatorio Nacional, Rio de Janeiro, Brazil, 1913. Aeronautisches Observatorium, Lindenberg, Germany, 1913. Italian Indo-Asiatic Expedition, 1913. Returned to Washington, 1916. Repaired, 1919. Sent to Observ- tory, Helwan, Egypt, 1920. Aeronautisches Observatorium, Lindenberg, Germany, 1913. Observatorio Astronomico Nacional, Tacubaya, Mex- ico, 1913. Landwirtschaftliches Institut, Moscow, Russia, 1914. Meteorological Observatory, Teneriffe, Canary Islands, 1914. Do. I4 ABBOT SILVER-DISK PYRHELIOMETER—ALDRICH 9 NO. Instru- Present ment constant SO 20 sre exes 0.3741 SU eee ee 0.3679 SR ee 0.36039 Sree 20 602 0'3074. Sal BE oeene 0.3622 Salees ieee 0.3748 SH ae 2 ee 0.3691 Subeskicosonn 0.3755 Sil. sae. 0.3730 Ss 35-2066. 0.3048 SAW paces 0.3640 Seg a78 cater 0.3648 Sala teeaes 0.3726 Sella Cesc 0.3764 Sullo ZOs eons 0.3553 Sip biome eee 0.3697 Sul 46 conse 0.3079 Syl ley. eee 0.3735 (Sip LV. agree 0.3793 So! ev eee 0.3658 SHAG See one 0.3705 BNE AT es scvccc 0.3735 BASS of asc 0.3827 Se Ost. ess 0.3586 Spec eee 0.3800 Sule Giivepcne 0.3755 Sal52 decuies 0.3787 Location and remarks Meteorological and Geophysical Service, Batavia, Java, 1914. Modernized and returned, 1947. Manila Observatory, Manila, P. I., 1915. Meteorological Office, London, England, 1915. Calama, Chile, Smithsonian South American Expedi- tion, 1918. Do. Jewish Consumptives Relief Society, Edgewater, Colo., 1920. Repaired, modernized, and sent to Carnegie Institution, Colorado and California, 1931. Mount Harqua Hala, Ariz., Smithsonian Solar Ob- serving Station, 1920. Reale Osservatorio Astronomico di Capodimonte, Naples, Italy, 1921. Riverside College Observatory, Sydney, Australia, 1921. Turned over to Commonwealth Observatory, Mount Stromlo, 1946. Do. Institutul Meteorologic Central, Bucharest, Roumania, 1922. University of Lemberg and Warsaw, Lemberg, Po- land, 1922. Argentine Meteorological Service, La Quiaca, Argen- tina, 1924. Do. Institut de Physique du Globe, Paris, France, 1924. Observatory on Zugspitze, Germany, 1924 (loan). Modernized, 1927. Sent to Institute of Meteorology, Nanking, China, 19209. Mount Harqua Hala, Ariz., 1923. Modernized, 1932, and sent to Carnegie Institution, Colorado and Cali- fornia, as substitute for S.I. 16. Stellenbosch University, South Africa, 1924. Meteorological Service, Rio de Janeiro, Brazil, 1925. Modernized and returned to Brazil, 1928. University of Aberdeen, Scotland, 1925. Meteorological Bureau, Riga, Latvia, 1925. Smithsonian Solar Observing Station, Mount Bruk- karos, Southwest Africa, 1926. Smithsonian Solar Observing Station, Calama, Chile, 1932. Note.— No. 47 and No. 48 have special 80-cm. tubes. Do. Battle Creek College, Mich., 1927. Physikalisch-Meteorologisches Observatorium, Davos, Switzerland, 1927. Repaired and returned to Davos, 1934. Meteorological Office, Simla, India, 1927. University Observatory, Kiel, Germany, 1927. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS Tir fe) Instru- Present ment constant Silerggeoteee 0.3797 Suis 54s 0.3824 SHG hsticnee 0.3848 Se Se (cleseanin 0.3742 SH Gy pei 0.3818 Sue S8enasee 0.3831 SaBeCpenoanc 0.3851 Sells OOsanenc 0.39045 Sa OMS asioc 0.3867 Silg62scene: 0.3960 Sue O3ee 0.3821 See 64) scans 0.3870 Sil 65822)0/46 0.3913 Sil Osasane 0.3868 SHE Cyfsococt 0.3877 SAMG8e ace 0.3788 SHM60% 2h 0.3849 SFOs sia 0.3816 Sania 0.3737 SIL 72 shone 0.3820 Sui Wwaaavce 0.3789 SUA sUiggowa.c 0.3962 Sua /jacnaee 0.3847 SH One 0.3876 Sullbevesnee ule 0.3922 Se 78a acces 0.3926 SH GOs ator 0.3736 Silo GOs oucas 0.3776 Subs Sri see. 0.3871 Location and remarks Commonwealth Solar Observatory, Canberra, Aus- tralia, 1928. Repaired and returned to Australia, 1937. Commonwealth Solar Observatory, Canberra, Aus- tralia, 1928. Repaired and returned to Australia, 1939. University of Illinois, Department of Botany, Urbana, 1928. Universitats-Sternwarte, Kiel, Germany, 1928. Institut Scientifique Chérifien, Rabat, Morocco, 1929. Oporto Observatory, Portugal, 1929. University of Arizona, Tucson, 1930. American Society of Heating & Ventilating Engineers, Pittsburgh, Pa., 1931. Repaired and returned to Pittsburgh, 1945. Meteorological Service of Martinique, Fort de France, 1932. Do. Blue Hill Meteorological Observatory, Milton, Mass., 1932. Repaired and returned to Blue Hill, 1941. Laboratoire Actinométrique de 1l’Observatorie de Trappes, France, 1933. Institute of Meteorology, Nanking, China, 1933. Institut de Physique du Globe, Paris, France, 1934. Tananarive Observatory, Madagascar, 1935. Institut de Physique du Globe, Paris, France, 1935. Geophysical Observatory, Pilar, Argentina, 1936. Do. Commonwealth Solar Observatory, Canberra, Aus- tralia, 1936. Repaired and returned to Australia, 1937. Commonwealth Solar Observatory, Canberra, Aus- tralia, 1936. Repaired and returned to Australia, 1939. Eppley Laboratories, Newport, R. I., 1937. New thermometer inserted and returned, 1939. University of Minnesota, Minneapolis, Minn., 1937. R. Dvorak, Agent, Prague, Czechoslovakia, 1938. Observatorio Central Meteoroldgico, Lisbon, Portugal, 1940. Geophysical Institute, University of Coimbra, Portu- gal, 1940. U. S. Weather Bureau, Washington, D. C., 1944. Physics Institute, Helsinki University, Finland, 1946. Hebrew Technical College, Haifa, Palestine, 1947. Institute for Advanced Learning, Dublin, Ireland, 1949. NO. I4 ABBOT SILVER-DISK PYRHELIOM ETER—ALDRICH iti Various A.P.O. instruments Instrument Constant Location and remarks POPEOL VV: oeaces:s 0.5118 Copper disk. 1906. Mount Wilson, Calif. JN eA OS AO Ley 0.5072 Copper disk. 1906. Mount Wilson, Calif. AA JEHO) WAUULS 5 ha5e 0.5150 Copper disk. 1906. U. S. Weather Bureau and Mount Wilson. EPO 8) aueis sistne oc 0.3760 First silver-disk instrument. 1909. Washing- ton and Mount Wilson. Repaired 1910 and called A.P.O. 8n1s. AEA)? Spite rene os ne 0.3786 Modernized 1927. Reserved at Washington for comparisons only. PE PUOV Oa aie 0.3631 Built 1910. Used at Washington, Mount Wilson, Mount Whitney, Algeria, Arizona. Modern- ized 1928 and called A.P.O. Opts. PNAE Os Opianicciasicale 0.3684 Zentralanstalt fur Meteorologie und Geodyna- mik, Vienna, Austria, 1928. Jaks Ba G va (0 eae aa 0.3720 Built 1923. Sent to Harqua Hala and Table Mountain. INSP LOOT ie acres cee 0.3675 Sent to Mount Brukkaros, Southwest Africa, 1925. Modernized and sent to Mount St. Katherine, Egypt, 1033. Repaired 1938 and sent to Tyrone, N. Mex., Miami, Fla., and Table Mountain, Calif. ACPAOMAI2) feiccissics 0.3618 Table Mountain, Calif., 1928. INGE ORIIB sen eces.s 0.3803 Mount St. Katherine, Egypt, 1932. Rebuilt 1938 and sent to Tyrone, N. Mex., Miami, Fla., and Table Mountain, Calif. PORE QTAY ® Seten 0% 0.3854 Mount St. Katherine, 1932; Tyrone, N. Mex., Miami, Fla., Table Mountain, Calif. 1 " Pal iif oa cis ack I aT eae? bi 7 i ei eh Ue e 5 oad ' ¢ % y i a s i 4 f ‘Vad e ‘ Sow Pig wey, ie ey i a) s 797 he - aie Di gi. Ai ; el, Oe i ihe ul Vawy iv, rare: "PL Pane Whi \y Vy ‘ ‘ ur Ff oe rie are ca) oy ¥ i] / a mene A ihe ot _ _ a) Ae PAK: 6 ave , lige ul ; . j ; ip ih iv? Mai me is iia nd tyne \ ie Pm oe ‘mere SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 111, NO. 14, PL. 1 ABBOT SILVER-DISK PYRHELIOMETER, WITH LONG TUBE ADOPTED 1927 fi}.t = 3 ei Gun Peles = | a |, ea i] ‘SMITHSONIAN MISCELLANEOUS COLLECTIONS _ VOLUME 111, NUMBER 15 THE ROLL CAEL ORY: THE EROQUOIS CHIEFS A STUDY OF A MNEMONIC CANE FROM THE Bee SIX NATIONS RESERVE (WirH 12 Prates) | BY | WILLIAM N. FENTON Bureau of American Ethnology Smithsonian Institution co (Pustication 3995) ea asi: ‘CITY OF WASHINGTON a PUBLISHED BY THE SMITHSONIAN INSTITUTION * _ FEBRUARY 16, 1950 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 15 ime MOLE CALL OF THE IROQUOIS CRIBS A STUDY OF A MNEMONIC CANE FROM THE SIX NATIONS RESERVE (WITH 12 PLATEs) BY WILLIAM N. FENTON Bureau of American Ethnology Smithsonian Institution (Pusiication 3995) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION FEBRUARY 16, 1950 The Lord Baltimore Preas BALTIMORE, MD., U. 8. A. CONTENTS Page IIE OCMUCEIONIE ote creltapaeisicjauis 2 viens «gia sle-eld ou vin eisias hele Soe ema ee eee I shencancvOrmeNndnewaSphager occ. cceic dees oo edeee ee eee. POE alate. I MIMI em rT cut Merersee eter eeheve iors cnet tale Scars feo Yalvers le aun iaitel mateo, avsee autre atedievsire. aca ne II PAM ITIISELITTIESPE CIM EM epererreyencial cise ovomeusote erokevelorore re atetene Suelelatele svete eer: 18 Weatetrial@descniptiotionrectericcis ceiecverte cccclsee eieiaie tone forcrsueie ve sth see eee 22 Shapesandseceneraleappearancesre ectes se se estee cee Goo 22 Miatenialmcancdecaimensionsmectrise. cies cise aac on =e os cee oles emete 59 shew Oneidar PieCtOPTApliSisacs.co eis Soe sien mi Davenhethon 4o |() by Wadondaherha 41 |()' » He Dea kae 42 {Q)| Be Skanyadariyoh Sees © Shadekaronyos | | ShakenJjohvaneh @ Kanokareh JPN, lo ——— Deshayenah SRiO Shodyenavat > ° Kanonkerihdayih = Deyohninhobhakaravenh bi Fic, 3—Front of the Condolence cane, The roll call of the Founders of the League. ONONDAGA | 19 DehadodaMtho? 20 one? asthene 21 Dehatga*aonts 22 Honya?daaji‘4ak 23 Awe? gen‘hyat 2k Dehaya’tgvar‘e? 25 Hononvie‘ai¢ 26 Govennen? shen’ donk 27 Ha‘hit hon? 28 Hoyonnyen‘ni* 29 Shodegva’’sen? 30 S*hagogent he? 31 e’ak‘ha“ wit 32 Sganawa' ai° Dega’ en‘ yon* Ged J1” nondave' het Gadagva"’aji? Shoyon’vea Eadya’ gen’ ne? Deyoen‘hyon’ got Deyot howe. gvi® Diyaven‘ het‘ hon? Fadondat he* ha? Desga‘‘he’? 43 Ay 45 46 47 48 ug 50 SENECA Skanyadai “iyo? Sha? degaen‘ hyes Shagen? djo“wane” Ge? no"gai® Nie‘ hanye’non’t Sadye‘navat Ganon‘ goi?da' wi? Deyonin‘ hogs’ven" NO. I5 ROLL CALL OF IROQUOIS CHIEFS—FENTON 69 is laid out in kernels of corn. The pictographs, with certain exceptions which resemble early treaty signatures and some war records of the seventeenth century, appear modern and give evidence of having been worked over and embellished in later times. Certain of them represent objects of recent historic introduction, a type of log house built at Grand River formerly (13), the nineteenth-century claw hammer (29). They are in character with the drawings of Chief Abram Charles and could almost have been made under his direction. Chief Charles was an advocate for the position that there were only 49 chiefs, that the Onondagas had 14 offices held by 13 persons. So was Andrew Spragg. The cane appears to have been equipped with 50 pegs, including 14 for the Onondagas. The last of these has been cut off, which means that the specimen antedates this controversy, for which we may take the year 1923, when it is first mentioned in Hewitt’s notes. Besides the alteration of the pegs, the cane shows the marks of successive owners. The possibility has been indicated that the picto- graphs were added to and altered or worked over by several persons who must have had the specimen in their possession for some time, probably as keepers. A. Spragg went so far as to inscribe his name, and we know that he was its last custodian. John Smoke knew of three holders—Ganawado who made it, Sheriff Billy Wage, and Spragg. This takes it back to about 1850, with but one weak link in the chain. Hale, who saw Billy Wage in action in the summer of 1883, does not say that he carried a cane, nor does he describe the staff carried by Silversmith who took over the role inside the long- house. Allowing 20 years apiece for each of the holders (and this is generous) we reach the year 1860 as the probable date of its manu- facture. This would be in the second or third generation on Grand River. Other evidence from related specimens shows that it combined several ancient ideas. Counting with kernels of corn in diagrams permeates Iroquois ceremonialism. The use of canes as symbols of age status has also been remarked. And we have shown that similar tallies and pictographs had earlier cultural beginnings. What we have then in the present specimen is a synthesis of several cultural traits which have long historic roots in the past. CONCLUSION In this study of the Condolence cane in the Cranbrook Institute we have tried to show its provenience, to date it within reasonable probability, to uncover its cultural roots, to explain its function 7O SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III within the Condolence Council, and to interpret its symbolism. It is certain that Andrew Spragg was a late holder of the ritual and keeper of the cane, which belonged in a way to all the Cayugas. Made some- time after the middle of the nineteenth century, probably about 1860, its use was confined to the Lower Cayuga band on the Grand River. Alterations in the specimen reflect local opinion on the Six Nations Reserve as to how many federal chiefs founded the League, how they were arranged, and the order of the roll call. Unquestionably it was devised on an ancient mnemonic design in order to preserve the memory of the Eulogy chant and roll call and to support the performance of the Condolence Council for installing new chiefs after settlement on the Grand River. Although it is now a sacred relic, and it has been reproduced once more by the Cayugas, the Cranbrook specimen does not antedate the American Revolution and the dissolution of the League in New York. It cannot be referred, therefore, to the period of the founding of the League. The investigation has forced us to study the institution of the Condolence Council. It has succeeded in unraveling the meaning of the pictographs and it has demonstrated how they relate to two local versions of the Eulogy chant—Newhouse (1885) and Charles (1917). It poses certain problems of kinship for future field work, since we now have before us the complete organization chart of the government of the Iroquois Confederacy, which was, perhaps, the classic example of the kinship state. BIBLIOGRAPHY ABERLE, S. D. 1948. The Pueblo Indians of New Mexico: their land, economy and civil organization. Amer. Anthrop. Assoc., Mem. No. 7o. BEAUCHAMP, WILLIAM M. 1905. Aboriginal use of wood in New York. New York State Mus. Bull. 89. Albany. 1907. Civil, religious and mourning councils... . New York State Mus. Bull. 113. Albany. Boye, Davin. 1906. The making of a Cayuga chief. Ann. Arch. Rep. Ontario, 1905, pp. 56-58. Toronto. BRINTON, DANIEL G. 1885. The Lenapé and their legends, with the complete text and symbols of the Walam Olum. Philadelphia. Bruyas, R. P. JAMEs. 1862. Radices verborum Iroquaeorum (Radical words of the Mohawk lan- guage). Shea’s Library of American Linguistics, No. 10. Cramoisy Press, New York. NO. 15 ROLL CALL OF IROQUOIS CHIEFS—FENTON Ais Bunze., RutH L. 1932. Introduction to Zufii ceremonialism. 47th Ann. Rep. Bur. Amer. Ethnol., pp. 467-544. CATLIN, GEORGE. 1848. Notes of eight years’ travel and residence in Europe. ... 2 vols. New York. CHADWICK, E. M. 1897. The people of the Longhouse. Toronto. CLARKE, Noau T. 1931. The wampum belt collections of the New York State Museum. New York State Mus. Bull. 288, pp. 85-121. Albany. CoLpEN, C. 1922. The history of the Five Indian Nations of Canada. ... 2 vols. Allerton Book Co., New York. DeVries, Davip PETER. 1857. Voyages from Holland to America, 1632-1644. New York Hist. Soc. Coll. (2), vol. 3, pt. I, pp. 9-136. FENTON, WILLIAM N. 1940. Problems arising from the historic northeastern position of the Iroquois. Smithsonian Misc. Coll., vol. 100, pp. 159-251. 1941. Iroquois suicide: a study in the stability of a culture pattern. Bur. Amer. Ethnol. Bull. 128, Anthrop. Pap., No. 14, pp. 79-137. 1944. Simeon Gibson: Iroquois informant, 1889-1943. Amer. Anthropolo- gist, vol. 46, pp. 231-234. 1946. An Iroquois Condolence Council for installing Cayuga chiefs in 1945. Journ. Washington Acad. Sci., vol. 37, No. 4, pp. 110-127. 1949. Seth Newhouse’s traditional history and constitution of the Iro- quois Confederacy. Proc. Amer. Philos. Soc., vol. 93, No. 2, pp. 141-158. Philadelphia. FLANNERY, REGINA. 1939. An analysis of coastal Algonquian culture. Catholic Univ. Amer., Anthrop. Ser., No. 7. Washington. ForEMAN, GRANT. 1946. The last trek of the Indians. University of Chicago Press, Chicago. FRACHTENBERG, LEo J. 1913. Contributions to a Tutelo vocabulary. Amer. Anthropologist, n.s., vol. 15, pp. 477-479. GALLOWAY, WILLIAM ALBERT. 1943. A sacred slab of the Shawnee prophet. Cranbrook Inst. Sci., News Letter, vol. 13, No. 1, pp. 6-7. Bloomfield Hills, Mich. Hate, Horatio. 1883. The Iroquois book of rites. Brinton’s Library of Aboriginal Ameri- can Literature, No. 2. Philadelphia. 1895. An Iroquois Condolence Council: a study of aboriginal American society and government. Trans. Roy. Soc. Canada, 2d ser., 1895-96, vol. 1, sect. 2, pp. 45-65. HECKEWELDER, REv. JOHN. 1876. History, manners, and customs of the Indian nations... . Mem. Hist. Soc. Pennsylvania, vol. 12. Philadelphia. 72 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Hewitt, J. N. B. 1920. A constitutional league of peace in the stone age of North America. ... Ann. Rep. Smithsonian Inst. for 1918, pp. 527-545. 1944. The Requickening address of the Iroquois Condolence Council. (W. N. Fenton, ed.) Journ. Washington Acad. Sci., vol. 34, No. 3, pp. 65-85. Hewirrt, J. N. B., and Fenton, Witt1aM N. 1945. Some mnemonic pictographs relating to the Iroquois Condolence Council. Journ. Washington Acad. Sci., vol. 35, No. 10, pp. 301-315. HorrMAN, W. J. 1891. The Mide’wiwin or grand medicine society of the Ojibwa. 7th Ann. Rep. Bur. Amer. Ethnol., pp. 143-300. Hunt, Georce T. 1940. The wars of the Iroquois. ... Madison, Wis. Jenness, D. 1932. Indians of Canada. Nat. Mus. Canada, Bull. 65, Ottawa. 1933. Three Iroquois wampum records. Nat. Mus. Canada, Ann. Rep., 1931, pp. 25-29. Ottawa. Jesuir RELATIONS. See THWAITES, REUBEN G. (Editor). Kou, J. G. 1860. Kitchi-Gami. Wanderings around Lake Superior. London. Laritau, J. F. 1724. Moeurs des sauvages amériquains. 2 vols. Paris. MALLERY, GARRICK. 1893. Picture-writing of the American Indians. t1oth Ann. Rep. Bur. Amer. Ethnol. Mooney, JAMES. 1896, The Ghost-dance religion and the Sioux outbreak of 1890. 14th Ann. Rep. Bur. Amer. Ethnol., pt. 2, pp. 641-1110. 1898. Calendar history of the Kiowa Indians. 17th Ann. Rep. Bur. Amer. Ethnol., pt. 1, pp. 120-444. Morean, L. H. 1878. Ancient society. Henry Holt, New York. 1881. Houses and house-life of the American aborigines. Contr. North Amer. Ethnol., vol. 4. 1901. The League of the Ho-de-no-sau-nee, or Iroquois. (Rochester, 1851). (H. M. Lloyd, ed.) 2 vols. Dodd, Mead Co., New York. O’CALLaGHAN, E. B. 1849-1851. The documentary history of the State of New York. 4 vols. Vol. 4, 1851. Albany. Parker, A. C. 1916. The constitution of the Five Nations. New York State Mus. Bull. 184. Albany. PARKMAN, F. 1851. History of the conspiracy of Pontiac. . . . Boston. Parsons, E. C. 1932. Isleta, New Mexico. 47th Ann. Rep. Bur. Amer. Ethnol., pp. 193- 466. RUSSELL, FRANK. 1908. The Pima Indians. 26th Ann. Rep. Bur. Amer. Ethnol., pp. 3-390. NO. 15 ROLL CALL OF IROQUOIS CHIEFS—FENTON 73 Sapir, EDWARD. 1913. A Tutelo vocabulary. Amer. Anthropologist, n.s., vol. 15, pp. 295- 297. SAVERY, WILLIAM. 1844. A journal of the life, travels, and religious labours of. . . . London. ScHootcraFT, H. R. 1845. Oneota, or characteristics of the red race of America. New York and London. SEAVER, JAMES EVERETT. 1932. A narrative of the life of Mary Jemison, the white woman of the Genesee. (C. D. Vail, ed.) Amer. Scenic and Hist. Preservation Soc. New York. SHEA, J. G. (Editor). 1880. A description of Louisiana . . . by Louis Hennepin. New York. Speck, F. G. 1942. The Tutelo spirit adoption ceremony. . . . Pennsylvania Hist. Comm. Harrisburg. Sprer, LESLIE. 1933. Yuman tribes of the Gila River. Univ. Chicago Press, Chicago. SWANTON, JOHN R. 1928. Aboriginal culture of the southeast. 42d Ann. Rep. Bur. Amer. Ethnol., pp. 673-726. 1946. The Indians of the southeastern United States. Bur. Amer. Ethnol. Bull. 137. TuHwaltTes, REUBEN G. (Editor). 1896-1901. The Jesuit Relations and allied documents. 73 vols. Cleveland. 1903. Father Louis Hennepin, a new discovery of a vast country in America. 2 vols. Chicago. VOEGELIN, ERMINIE W. 1939. Culture parallels to the Delaware Walam Olum. Proc. Indiana Acad. Sci., vol. 49, pp. 29-31. 1942. Notes on Ojibwa-Ottawa pictography. Proc. Indiana Acad. Sci., vol. 51, pp. 44-47. WaArbLe, H. NEWELL. 1949. Janus-faced pipes. Pennsylvania Archaeologist, vol. 10, pp. 27-20. Waucu, F. W. 1916. Iroquois foods and food preparation. Canada Dep. Mines, Geol. Surv., Mem. 86, Anthrop. Ser., No. 12. Ottawa. Wulitt, Lestiz A. 1932. The Acoma Indians. 47th Ann. Rep. Bur. Amer. Ethnol., pp. 17-1092. WitLoucuey, C. C. 1938. A Mohawk (Caughnawaga) halter for leading captives. Amer. Anthropologist, n.s., vol. 40, pp. 49-50. ZEISBERGER, DAvIp. 1910. History of the northern American Indians. .. . (1779-80). (Archer Butler Hulbert and William Nathaniel Schwarze, eds.) Ohio Archeol. and Hist. Soc. Quart., vol. 19, pp. 1-189. Columbus, Ohio. PLATES - SMITHSONIAN MiSCELLANEOUS COLLECTIONS Voli Non tS Pl ot 1. GRAVE OF PATTERSON SPRAGUE IN LOWER CAYUGA CEMETERY, SIX NATIONS RESERVE (R. T. Hatt photograph. ) 2. A LOG HOUSE AT SANDY’S CORNERS, SIX NATIONS RESERVE, ONCE OCCUPIED BY ANDREW SPRAGG IROQUOIS CHIEFS ne of Andrew Spragg. re C Cayuga Condolence W ae k WL O =] aa < O | pa O & W I k v SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 15, Pl. 3 Ren Se wert | CHIEF’S CANES FROM THE SIX NATIONS RESERVE (National Museum of Canada photograph. ) SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 15, Pl. 4 Re a Me Si = & S Bree C CHIEF’S CANES FROM CANADA a and b, Janus-faced cane from Chiefswood collection (Royal Ontario Museum of Archaeology ) ; two sides showing face. c and d, Canes from the Huron or Iroquois of eastern Canada (Quebec Provincial Museum). SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 15, Pl. 5 a CANES WITH CLAN EFFIGIES, AND A FALSE-FACE a, Bear effigy; b, Wolf effigy; c, False-face or masker; d, the cane of the 12th Onondaga chief S‘agogen’he’ (No. 30 on the cane). (Milwaukee Public Museum photographs. ) SMITHSONIAN MISCELLANEOUS COLLECTIONS Vols i Nos 25) Pline 1. DETAIL OF CLAN EFFIGIES ON PLATE 5 a, Bear; b, Wolf. (Milwaukee Public Museum photographs. ) 2. DETAIL OF CLAN EFFIGIES ON PLATE 5 a, False-face; b, Wolf. (Milwaukee Public Museum photographs.) SMITHSONIAN MISCELLANEOUS COLLECTIONS Voli ete Nos 157 sPlis CANE USED AT A CHIEF’S WAKE, SIX NATIONS RESERVE a, Over-all view of cane (collected by J. N. B. Hewitt, 1010; U.S.N.M. No. 384288, Division of Ethnology); b, front view detail: notice of death and the funeral; c, right side detail: wampum belts of notification; d, left side detail: passage of the soul to the hereafter. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol. 111, No. 15, Pl. 8 PICTORIAL RECORD STICK OF THE SENECA PROPHET’S REVELATION (Milwaukee Public Museum photograph. ) SMITHSONIAN MISCELLANEOUS COLLECTIONS Vol 1 N 5 dina [Mey ss fells © } PA : od “ ugtng ) a , = j 1 2. G71 yea Ae he ha, / : ra " Ley Ue LAA ! rive ins cre {dg / meri tc Userhye « ! , Le Aa Fe oh At nh, 7 Fierce, & SNe °< § @eak a Ly nh wa thah, 2 umn = ; ; ‘ oe rrfea he Aa Pina deh, 3 eg ff as rs eo. * bea WES “4 , ‘ + « g - “ee 7 in@ > §Sha ver. ho wa neh, Up Xm o& ‘ . hele @x De youn he gwertk, eZ ~~ ag @a @x ss a. 4 he NS O Ki Fe go wah . ¢ # + Bt oeg , c ees ; : .s Dehkerinahavinch FJ Oe ue ~ 8 yok ka slainen se vont hak, ¥ - : Ges. 4 vot 4 3 $2 Bio shoaronwanch .« Y Sot oan ane ae eh ae 285 ° 4 ‘ \ Ee GC Ove n aol de ale? ~* vA 1b- O dats te deh, g. WN kano gen é yah, Ti Di Yo ha-gwen deh. /3\> : NR ay Oho nol SES, K-. /¥ De ho nave ken, be 15 ? [ dyu don 71€Qa tha, = 4 ~ SETH NEWHOUSE’S ROLL OF THE CHIEFS (CA. 1885) The Mohawk and Oneida rosters with mnemonic. SMITHSONIAN MISCELLANEOUS COLLECTIONS Volk Tes Ss NonetS) Plea, vA Kh 2 dahon dea yenh, TT Ra nya das ha youh, [S$ Ron wats ha dor honk. pe > : ( Ovnon dn gela ka.) ed do darhenh, [9 ed wen ni sera, 20. De hat bn dane OE Va dayji iva hens, ae Olh i Ae tg ¢ Deka yal gia veh ah * 5 os 3 Onon wi rhb" 2 x 7 Oe wei 7% 3€ Pon ni, 26 ¢ J viv honk. 27 /0 bi Ve iva yonh nyca nih, 2 Xd. Thole cle gGwa sth, 29. oo4 Siu Ao he ae . 20 aN SETH NEWHOUSE’S ROLL OF THE CHIEFS (CA. 1885) The Onondaga roster. . SMITHSONIAN MISCELLANEOUS COLLECTIONS Walls abril Ney aby Tei5 alal Bi : 4 ; , E. Atte rr { Ps ¥ 4 Sha vas wa di. | ds ( A 074 Yyouk gon hak Pet) Kyha-ea yoih, SA Li. PE: nee 7 He Pates Ray: ee ha a givaya ee as a Shor DS oun wea ate Apt aise Wa lya se ron neh BY 4 a De ae ron yor koh, I i 2/4 —— De yot ho veh gwenh, JF MW by. we rtf... Da 1s | Wa don da op ta, 4.f os uf? & Deshae . 42 : J ae SETH NEWHOUSE’S ROLL OF THE CHIEFS (CA. 1885) The Cayuga roster. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vo)s Ti, (Nox 5) 3P i 2 ie Sha Des agri ‘vt yok, 43 Nae oe S 2 Sha de ha von YES, 44 - “np. . Sha hen goh wa uk 4s ie nets BE ka pio ha rh AG os Vir Des ha Ye nah, 47 | Oho dye aoa wat, G5 Ka non he rth da wih, AG De yok nin har ha ha'vawenh.) 50. or SETH NEWHOUSE’S ROLL OF THE CHIEFS (CA. 1885) The Seneca roster. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 16 ine FORMS OF THE BLACK HAWK-EAGLE (WitH ONE PLATE) BY HERBERT FRIEDMANN Curator, Division of Birds, U.eS. National Museum 7 Wie . OXLOS bo TITV Te Aingro™® (Pus.ication 4013) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION FEBRUARY 28, 1950 The Lord Baltimore Press BALTIMORE, MD., U. 8 A. THE FORMS OF THE BLACK HAWK-EAGLE By HERBERT FRIEDMANN Curator, Division of Birds, U. S. National Museum (WitH ONE PLATE) The black hawk-eagle, Spizaetus tyrannus (Wied), has always been considered as a species with no geographic races, and, indeed, I so assumed it to be a number of years ago when working on this genus for the eleventh volume of Ridgway’s “Birds of North and Middle America.” However, recently I have had occasion to examine far more extensive material than was formerly available, and I find, quite contrary to my earlier opinion, that there are two morphologically separable geographic units in the species. The type of tyrannus came from Ilha do Chave, below Quartel dos Arcos, Rio Belmonte, Bahia, Brazil, and the only other names applied to the species are likewise based on eastern and southeastern Brazilian birds—Harpyia braccata Spix (Avium species novae .. . vol. 1, p. 7, pl. 3, 1824 (=1825)) from Sao Paulo, Brazil, and Spizaetus spixii Des Murs (Rev. Zool., vol. 10, p. 325, 1847) which is merely a new name for H. braccata Spix. Birds from México and Central America, and from northern and western South America, south to Para, the Amazon Valley, and west- ern Brazil (Rio Purtis and Rio Jurua and Mato Grosso) and Bolivia are different from specimens from Bahia, Espirito Santo, Rio de Janeiro, Santa Catharina, Minas Gerais and Sao Paulo in eastern and southeastern Brazil, and for them is proposed the name Spizaetus tyrannus serus, new subspecies Type.—U.S.N.M. No. 206391, ad. 3, collected at Rio Indio, near Gattin, Canal Zone, Panama, March 4, 1911, by E. A. Goldman (orig. No. 13928). Subspecific characters——Similar to the nominate race but with the flanks and thighs more heavily marked with white cross bars, and with the under wing coverts much more whitish, less blackish ; these feathers white, rather sparingly marked with blackish in serus, and almost wholly black in tyrannus. Size of serus averaging smaller than the nominate race. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 16 i) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Description of type.—All feathered parts of top of head black, the feathers of the crown and the occipital crest with their basal half or more pure white, much of this color usually showing through among the overlapping feathers ; scapulars, interscapulars, back, upper wing coverts black ; remiges externally fuscous black to black, crossed by five or six dull fuscous bars about equally spaced, the bars being slightly narrower than the black interspaces and becoming mottled or edged with whitish on the inner margin of the inner webs, these bars grayish white on the under surface of the remiges; rump black, the upper tail coverts black, each feather crossed by two or three nar- row white bars which are slightly mottled with brownish gray, and also narrowly tipped with white; rectrices black paling to whitish basally and crossed by four broad bands of deep ashy gray mottled with pale fuscous and whitish, and narrowly tipped with whitish; all the tail bands white on the under surface; chin black, the feathers white basally; feathered portions of lower cheeks and auriculars, throat, breast, sides, and upper abdomen black; lower abdomen and flanks black spotted sparingly with narrow, semitransverse white flecks ; thighs black broadly barred with white, the black interspaces somewhat broader than the white bars (in typical tyrannus these white bars are very narrow lines) ; under tail coverts like the thighs and flanks but with the white bars still broader and more broadly spaced ; under wing coverts mostly white broadly barred with black, the outer, greater, under primary coverts white with only a sub- terminal and a subbasal black bar (in typical tyrannus the under wing coverts are black with sparse and narrow white barrings). Young birds of both races tend to have more white on the under wing coverts and tibiae than do the adults, but even here the difference between comparable birds of the two subspecies is striking. Measurements of type-—Wing 381; tail 309.8; culmen from cere 27.5; tarsus 81 ; middle toe without claw 44 mm. In size S. ¢. serus averages smaller than S. t. tyrannus; thus males of serus have wing lengths of 354-394 (379.5 mm.) and tails 291.1- 325 (309.8) in length as opposed to wing lengths of 362-422 (398.3 mm.) and tail lengths of 365-367 (366 mm.) in the nominate race; females of serus have wing lengths of 353-444.5 (412.5 mm.) and tail lengths of 289-386 (337 mm.), while females of tyrannus have wings of 428-460 (443.7 mm.) and tails measuring 370-405 (388.8 mm.) Range.—Resident in heavily forested areas of the tropical zone from southern México (Oaxaca, Chiapas, San Luis Potosi, Puebla, Veracruz, Campeche, and Yucatan) south through Guatemala, Hon- NO. 16 THE FORMS OF THE BLACK HAWK-EAGLE—FRIEDMANN 3 duras, British Honduras, Nicaragua, El Salvador, Costa Rica, and Panama to Colombia, Ecuador, Peri, Venezuela, Trinidad, the Guianas, northern and western Brazil (Para, the Amazon Valley, Rio Purts, Rio Jurua and Mato Grosso), to Bolivia (Santa Cruz). The nominate form appears to be restricted to eastern Brazil (Bahia, Espirito Santo, Minas Gerais, Rio de Janeiro, Santa Catha- rina, and Sao Paulo) and possibly ranges to northeastern Argentina (Misiones). The species has been recorded once from Paraguay (Sapucai), but, in the absence of material from that country, it is not possible to say which race may have been involved. Material examined.—Thanks to the cooperation of the authorities of the Chicago Natural History Museum and the Museum of Com- parative Zodlogy, I have been able to add to the material in Wash- ington and have thus personally studied 30 specimens. In addition to these, I am able to include here notes on a still larger number of specimens in other museums, kindly made for me by Dr. Pinto on the birds in the museum at Sao Paulo, Brazil; by Mr. Peters on addi- tional specimens in the Museum of Comparative Zoology, Cambridge ; by Mr. Todd on those in the Carnegie Museum, Pittsburgh; by Mr. deSchauensee on the birds in the Academy of Natural Sciences of Philadelphia; and by Dr. Zimmer on those in the American Mu- seum of Natural History, New York. The combined material may be listed as follows: Spizaetus t. tyrannus—Eastern and southeastern Brazil (Bahia south to Sao Paulo) 20 (including the type) ; “South America” 2; total 22 specimens. Spigaetus t. serus—Mexico 8; Guatemala 1; British Honduras 1 ; Honduras 1; Nicaragua 1; Costa Rica 4; Panama 10 (including the type) ; Colombia 4; Venezuela 1; Dutch Guiana 6; Ecuador 2; north- ern and western Brazil (Para, the Amazon Valley, Rio Purts and Rio Jurua, Mato Grosso) 9; Bolivia 2; total 50 specimens. Remarks.—The two species of Spizaetus inhabiting the American Tropics each divide into two races, but the geographic pattern of this division is quite dissimilar in the two. In the case of the present species the data, still incomplete for much of the interior of Brazil (states of Maranhio, Ceara, Parahyba, Goyaz, southern Para, and Mato Grosso), indicate that the nominate form is largely a bird of the coastal or semicoastal forested areas from Bahia southward to Sao Paulo, and that the species apparently does not occur in the adjacent (to the west) extensive areas of “campos” country of the “Planalto” and of the still farther inland grasslands stretching from south of the Amazon forest to parts of northern and eastern Bolivia. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS fit The race serus ranges from the Amazonian forest and its outliers (from Para in the east, and in the west from the Rio Purts and Rio Jurua, from Mato Grosso (Sao Luiz de Caceres) and from Santa Cruz in Bolivia) northward to Pert, Ecuador, Colombia, Surinam, Venezuela, Central America, and México. In Spizaetus ornatus, which, as a species, seems to have a range geographically and ecologically very similar to that of S. tyrannus, the northern race vicarius occurs from southern México, south across Central America to Colombia west of the eastern Andes, and western Ecuador, possibly to Pert, while typical ornatus is found from Vene- zuela and the Guianas southward across Brazil to Bolivia and Para- guay. In Spizaetus ornatus it looks as though the Andes constituted a possible barrier between the component forms, whereas in S. tyrannus this is not the case. WOT Silos t 1G . [IZe1g OoIxXoyy ‘sederyy Uto}Seo ‘O[JIAUTOf WOT, SHMUDAN, SUDAN, Snjpansid > 319 V3S-AMVH HOV1E 4 j \ fs i By et ee SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 17 Roebling Fund BeRIODIC INFLUENCES ON WASHINGTON AND NEW YORK WEATHER OF 1949 AND 1950 BY, € G, ABBOT Research Associate, Smithsonian Institution (PusticaTion 4015) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION MARCH 22, 1950 The Lord Waftimore Press BALTIMORE, MD., U. 8. A. % Roebling Fund PERIODIC INFLUENCES ON WASHINGTON AND NEW YORK WEATHER OF 1949 AND 1950 By,G/G. ABBOT Research Associate, Smithsonian Institution A. PRECIPITATION AT WASHINGTON In Smithsonian Miscellaneous Collections,t I have traced the in- fluence of a cycle of 27.0074 days on Washington precipitation. This cycle is thought to be associated with the rotation period of the sun. Dates were assigned when it was expected that, on the whole, greater average precipitation would fall in Washington than on the average of all other dates. In 1949, for the sixteenth consecutive year, this proved to be so. The ratio of average daily precipitation on pre- ferred dates of 1949 to average daily precipitation on all other dates was 1.56. Basic statistical study of the years 1924 to 1941, inclusive, indicated the expected ratio to be 1.42. The average ratio for 16 years ending with 1949 is 1.47. In detail the year 1949 yielded the following values : TABLE 1.—Statistics of Washington precipitation, 1949 (Values in inches) Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Year pe ee | Ed. Saisie 0.219 0.143 0.207 0.098 0.229 0.044 0.070 0.269 0.209 0.024 0.025 0.067 0.137 ‘per day § All other.. 0.118 0.090 0.053 0.040 0.139 0.077 0.220 0.000 0.039 0.0161 0.025 0.043 0.088 Ratio Si os eee 1.86 1.59 3.90 2.45 1.65 0.57 0.32 * 5:40) (0315, 1.00) 92.57; 1.56 iiotal ppt. ......... 5.08 3.21 3.96 2.0% 5.65 1.85 4.57 4-57 3-55 3-21 0:74) e720 ALIS Normal ppt. ....... 3.55 3-27 3.75 3-27 3-70 4.13 4.71 4.01 3.24 2.84 2.37 3.32 42.46 )Percent of normal .. 143 98 106 61 153 45 o7) i114. (10 Trg 31 2 94 * Infinitely large. Preferred days of 1949 had a higher average precipitation than all other days in all months but June, July, and October.’ 1 Smithsonian Misc. Coll., vol. 104, Nos. 3 and 5, 1944; vol. 110, No. 4, 1948; vol. 111, No. 5, 1949. 2In November recorded rainfall averaged the same in both groups, but traces of rain fell on 4 preferred days and on only 3 other days. During July and August rain fell copiously in northern Washington, and nearby, on preferred days when none was recorded at the Weather Bureau. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 17 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Table 2 gives the dates for 1950 when the average daily precipita- tion in Washington is expected to exceed the average daily precipita- tion in this city on all other days. In the first column are given in Roman numerals the day numbers of the 27-day cycle when higher TABLE 2.—Predicted dates for the year 1950 when average daily precipitation should exceed average daily precipitation of all other dates of the year 1950 in Washington, D. C. “Preferred” cycle places Jan. Feb. Mar. Apr. May June be renee re Cie II 7 6 2, 29 26 22 1 Derren ee 1 8 7 3, 30 27 23 Tes See setere oper 13 9 8 4 1, 28 24 |" SOR enc 14 10 9 5 2, 29 25 VE CARLES, Me 15 II 10 6 3, 30 26 SEs ee 22 18 177 13 10 6 SOE teense 23 19 18 14 II 7 XO) A accor 2 21 20 16 13 9 CVA ae 27 23 22 18 15 iti >, Q\Va) Ih bi Wiereeertoicat: 1, 28 24 23 19 16 12 DOG LI dh Bees reser 5 I, 28 27 23 20 16 RO IVIT. ae eee 9 5 4, 31 27 24 20 DOO VTi be sie ees 10 6 5 I, 28 25 21 “Preferred” cycle places July Aug. Sept. Oct. Nov. Dec. [fa oad 2 cpaet ac 19 15 II 8 4 1, 28 ITs seesre eae 20 16 12 9 5 2, 20 1D eesea operons 21 17 13 10 6 3, 30 DVRs oor 22 18 14 II 7 AySt Vitek ares 23 19 15 12 8 5 DSS) Scere reso 3, 30 26 22 19 15 12 RST a ake repsie ses 4, 31 27 23 20 16 13 XV is Sree 6 2, 20 25 22 18 15 VT Aon cee 8 AS Sr 26 23 19 16 OVA aerne eae 9 5 1, 28 25 21 18 3.69.4) Il PARA eative 13 9 5 2, 29 25 22 PREXS VIG a alae ela 17 13 9 6 2, 20 26 OVAL eyes ee 18 14 10 7 3, 30 27 precipitation is expected. These values arise from the statistical study, 1924 to 1941, above mentioned. The other columns give the actual days in the 12 months of 1950 when these Roman cycle dates will occur. In other words the remaining columns give the “preferred” dates for 1950. While it is expected that for the entire year 1950 the “preferred” dates will yield higher average precipitation than all others, and even that this will be so for most of the individual months of 1950, the probability that any individual “preferred” day will yield % — NO: L7 WASHINGTON AND NEW YORK WEATHER—ABBOT 3 precipitation is scarcely above 50-50. Recent press accounts of sur- prising accuracy in these predictions for individual days of past years refer merely to lucky happenings. The basic statistical tabulation from 1924 to 1941, to which I re- ferred above, and on which table 2 is based, began January 1, 1924. The length deduced for the cycle is 27.0074 days. In 352 cycles of this length there are 9506.6048 days. In the years 1924 to 1949, inclu- sive, there were 9497 days. Hence the Roman cycle date I falls on January II, 1950, as given in table 2, being 10 days later in January than the original Roman cycle date I, which fell on January 1, 1924. B. TEMPERATURE AT WASHINGTON In previous papers I have drawn attention to a regular periodic variation of 6.6485 days’ length in the output of radiation from the sun. Though quite regular intervals occur in the solar variation, terrestrial responses thereto are sometimes I, 2, or rarely 3 days from their expected dates. This is due to the complexity of the atmos- pheric constituents and reactions. All terrestrial responses to solar impulses are subject to lag. For instance, the warmest part of the day occurs from I to 6 hours after noon at various stations of the earth. The lag is not constant from day to day at any station. Hence, from analogy, the irregularity of terrestrial responses to the 6.6485- day solar variation is not surprising. Nevertheless they are notable in magnitude. As shown in earlier papers they range from 2° to 20° F. in the temperature of Washington. This statement will be found confirmed in figures 2 and 3. Notwithstanding the differences in lag just referred to, which cause displacements of the terrestrial responses, it seemed to me worth while, in January 1948, to predict for the ensuing year the 55 dates when minima of temperature with respect to surrounding days might be anticipated in Washington. In doing so I recognized that actual minima would sometimes fall 1, 2, or even rarely 3 days from the dates predicted. In January 1949 the prediction was compared with the event. Figure 1 shows the numbers of days when the observed minima coincided, or fell 1, 2, or 3 days from the predicted dates in 1948. The published predictions for 1948 and 1949 (above cited) were made with the original value of the length of the period. I now give in table 3 new dates to replace those published for 1949 in a previous 8 Smithsonian Misc. Coll., vol. 107, No. 4, 1947; vol. 111, No. 6, 1949; vol. 111, No. 13, 949. The period was originally determined as 6.6456 days. But in the last of the three papers cited a correction of 0.0029 days was found, thus making the preferred period 6.6485 days. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Fic. 1—Frequency of minima, —3 to +3 days from dates predicted in 1948. TABLE 3.—Corrected dates in 1949 and 1950 when minima were due in Washington temperatures 1949 18 20 19 14 18 20 17 19 14 18 20 16 25 27 31 28 31 30 hrwrob) AHP ANH DAH UU 1950 19 14 20 15 18 14 17 19 15 18 14 17 NOS L7, WASHINGTON AND NEW YORK WEATHER—ABBOT 5 paper. They are based on the corrected period, 6.6485 days, and as- suming January 17.0000, 1946, as the basic date of temperature minimum at Washington. In checking the results for 1949 I prepared a table of 7 columns and 55 lines. The departures from normal temperatures for dates nine Ht a a ae AAT A | NNT AE eit | iT eSNOK 3O DEC. 5 25% 20° 1Sin /o° WAS ae San Fic. 2—Temperature departures from normal at Washington, December 10949. Dotted lines indicate predicted dates for minima. predicted to yield minima of Washington temperatures according to table 3, I entered in the fourth column. In each of the 55 lines I then entered the departures for the 3 preceding and 3 following days, along with the central predicted date. This made up a table of 7 columns and 55 lines. In this way 20 days were used twice, owing to overlapping. To obviate this defect, I cut off the temperature de- partures on overlapping dates, alternately from the first and the seventh columns, removing 10 departures from each of these columns. From this table, which to save printing I do not publish, I found the column in each line which carried the minimum temperature for that 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES isk 20° (Si 10° 1025 NOK 30 OC. S /0 IS 20 es JO NEW YORK Fic. 3.—Temperature departures from normal at New York, December 1949. Dotted lines indicate predicted dates for minima. line, and then took the sums for each column. The results are shown in table 4. TaBLeE 4.—Frequency of minima of Washington temperatures with respect to the dates of predicted minima of 1949 Days from predicted dates......... —3 —2 —I (0) +1 +2 443 Numbers of days of minima........ 5 II AnerT 9 9 6 I also took the mean values of the departures of temperature from normal for the 7 columns of the table of 55 lines. These results are given in table 5. TABLE 5.—Average temperature departures at Washington with respect to the dates of predicted minima of 1949 Days from predicted dates......... —3 —2 —I fe) +1 +2 +43 Average temperature departures... 4°46 5°16 4°87 5°42 3°06 3°60 3°87 It is clear that the prediction for 1949 was less successful than that for 1948, illustrated in figure 1. In other words, the displacements NO. 17 WASHINGTON AND NEW YORK WEATHER—ABBOT 7 of temperature minima from dates expected to suit the solar period of 6.6485 days were more considerable in 1949 than in 1948. These displacements, as explained above, are attributed to the complexities of atmospheric reactions. I postpone further remarks to the “Dis- cussion” below. C. NEW YORK TEMPERATURES As shown in a previous paper,* a comparative study of Washington and New York temperatures over the interval of 21 years, 1928 to 1948, with reference to the solar period of 6.6485 days, indicated that this solar variation strongly affected the temperatures in both cities, and almost identically. By the kindness of E. J. Christie, Meteorolo- gist in Charge at New York, I have received Form 1030 for the year 1949, which gives the daily departures from normal temperatures there. I have treated these data exactly as I have described above for Washington. The results are given in tables 6 and 7. TABLE 6.—Frequency of minima of New York temperatures with respect to the dates of predicted minima of 1949 Days from predicted dates......... —3 —2 —I ) +7 +2 Numbers of days of minima........ 9 4 iit 13 4 -|- Cw TABLE 7.—Average temperature departures at New York with respect to the dates of predicted minima of 1949 Days from predicted dates....... —3 —2 -—I 0 5 -+--2) -3 Average temperature departures... 4°35 4°49 4°75 4°73 3°53 3°76 4°00 D. DISCUSSION OF TEMPERATURES Neither at Washington nor at New York does the year 1949 show, as a whole, a clear indication of the importance of the solar variation of 6.6485. Were it not for the extensive evidence given in Smithsonian Miscellaneous Collections, vol. 111, No. 13, one would conclude from the year 1949, alone, that this supposed periodic temperature effect is illusory. Certainly one would be inclined to conclude that the period offers little promise of being a useful means for predicting tempera- tures a year in advance, as was attempted with some success in 1948. But individual months, even of 1949, give a different impression. In figures 2 and 3 I give for Washington and New York the actual marches of departures from normal Fahrenheit temperatures, from November 25 to December 31, 1949. In both figures the dotted lines 4 Smithsonian Misc. Coll., vol. 111, No. 13, 1949. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. it are drawn exactly on the dates when minima of temperatures should arrive, counting intervals of 6.6485 days from January 17.0000, 1946. The December minima in these figures 2 and 3 all fall within 1 day or less of the expected dates. As shown in Smithsonian Miscel- laneous Collections, vol. 111, No. 13, so in figures 2 and 3, the two cities behave almost identically. The average range of the periodic fluctuation is 18°6 F. at Washington, and 18°o0 F. at New York. This is about at the maximum for the year. In July, as stated in Smith- sonian Miscellaneous Collections, vol. 111, No. 13, the range is much less. But surely one may conclude that, though not as yet thoroughly available for long-range temperature prediction, the period of 6.6485 days is a major factor in weather. _ SMITHSONIAN MISCELLANEOUS COLLECTIONS Soke VOLUME 111; NUMBER 18 (End of Volume) | * TREE GROWTH AND. RAINFALL— | A STUDY OF CORRELATION Bk. AND METHODS BY ' WALDO S. GLOCK | Macalester College Ie pects C : % Tee, a Pe » ih DK NG ie Wb 1 (ap lis vd ‘4 (Pustication 4016) Habe CITY OF WASHINGTON _ | |. PUBLISHED BY THE SMITHSONIAN INSTITUTION HC RaEN TES OCTOBER 25, 1950 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 111, NUMBER 18 (End of Volume) TREE GROWTH AND RAINFALL— PE COlUDY .OF CORRELATION AND METHODS BY WALDO S. GLOCK Macalester College (PusticaTion 4016) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION OCTOBER 25, 1950 TBe Lord Baltimore Press BALTIMORE, MD., U. & A. TREE GROWTH AND RAINFALL—A STUDY OF CORRELATION AND METHODS? By WALDO S. GLOCK Macalester College The purpose of the present study is to test critically the covariation of tree growth and rainfall—tree growth as measured by the thick- nesses of growth layers on increment cores. Three desiderata were the bases of the study. First, an altitudinal zone was to be selected above the region of violent fluctuations of soil moisture but below the region of excessive accumulation and possible carry-over from season to season. This would avoid the forest-border zone where temporary depletion of soil moisture during the growing season could bring growth to a halt temporarily and possibly cause multiplicity of growth layers during that season. Second, the number of trees sampled was to be held to a minimum in order to avoid excessive duplication of record and to avoid inclusion of trees from habitats so diverse that the merged record would become blurred. The number, however, was to be sufficiently large to absorb any differences in relative growth-layer thicknesses from tree to tree due to slight variations in site factors local to the individual trees. Third, the trees were to be selected in the field on the basis of ecologic principles, after which each core, unless marred by accident or disease, would enter into the group record whether or not the relative thicknesses of its growth layers closely agreed with those of the other cores. The writer is aware? of the shortcomings and the possible mis- representation inherent in the use of rain-gauge records taken some 1Grateful acknowledgment is made to Dr. A. Wetmore and to the Smith- sonian Institution, which supported the entire project. To Dr. R. Sidwell grati- tude is due for courtesies extended in the field. Herbert Gross, of Macalester College, was of much assistance not only in the preparation of the figures but also in the lively interest he evinced in the problem. Rainfall data from 1931 to 1046 were obligingly supplied by the Weather Bureau office in Albuquerque, N. Mex. 2 Bot. Rev., vol. 7, pp. 649-713, 1941; Journ. Forestry, vol. 40, pp. 614-620, 1942. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 111, NO. 18 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III miles from the site of the trees, in the use of a single radius to repre- sent the entire volume growth of a tree, and in the emphasis on a single growth factor. However, if significant results can be obtained, in spite of handicaps, by proper selection of trees from the correct habitat, a critical test is highly worth while in view of the simplicity and directness of method. Heretofore, many of the correlations * between tree growth and rainfall have been discouraging unless the data were smoothed to an extent that direct responses were masked and only general trends revealed. LOCATION AND TREE DESCRIPTION The increment cores came from trees that grew near and on Holman Pass, in the Sangre de Cristo Range of north-central New Mexico, about 41 miles by road or about 35.5 miles airline north-northwest from Las Vegas (fig. 1). In so far as the life zones were concerned, the collection extended upward from mid-Transition into the lower portion of the Canadian. All the trees sampled were dominant or codominant and in the timber stage of development. On the whole, the ponderosa pines were slightly more mature than the other species. Neighboring trees not sampled were sufficiently distant to avoid undue competitive in- fluence as far as site factors were concerned. Furthermore, the locations were chosen so that abnormal drainage toward or away from the trees was at a minimum. The soils were in no sense tight or lacking in aeration. In all, nine trees were sampled and designated by the initials HPC, for Holman Pass Collection. The trees from which samples HPC 1 to 4 were taken grew on a nearly flat area a mile southeast of the Pass at an elevation of 9,000 feet. All four were within 150 yards of one another. The black soil contained numerous pebbles and boulders. Cores HPC 5 and 6 came from trees that grew on the Pass itself at an elevation of 9,450 feet. In spite of the fact that the site was on top of the actual pass, the trees stood in the middle of a broad, essentially flat area. The soil was derived from shale and sandstone bedrock, fragments of which remained. Between 7 and 8 miles west of the Pass and down Rio Pueblo Canyon the location of 8 Many of these are listed and discussed in Bot. Rev., vol. 7, pp. 687-698, 705- 713, I94I. # Following the classification of James W. Toumey and Clarence F. Korstian, Foundations of silviculture upon an ecological basis, p. 268, 1937. New York. = No. 18 TREE GROWTH AND RAINFALL—GLOCK 3 Scale Of Miles 0 20 | 40 60 _ Fic. 1.—General location map for Holman Pass collection and rainfall stations. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOE. TEs the trees for HPC 7 to 9 was chosen at an elevation of approximately 8,000 feet. Here ponderosa pines were more mature and more domi- nant than at the other two sites. The trees grew on top of a very gently sloping terrace whose edge stood 20 feet above the stream channel. Toward the south the terrace top rose gently to a steeper, heavily wooded slope over 400 feet away. The soil, granitic in com- position, contained numerous pebbles and boulders, Individual tree and core descriptions are given below in concise form: HPC 1. Ponderosa pine (Pinus ponderosa). 14 inches DBH. Average thick- ness of growth layers 1.61 mm. Range 1850-1946. Growth-layer sequence vari- able. Average departure from mean 0.58 mm., or 36 percent. HPC 2. Ponderosa pine. 12 inches DBH. Distant 100 yards from HPC 1. Average thickness of growth layers 1.86 mm. Range 1850-1946. Growth-layer sequence variable. Average departure from the mean 0.54 mm., or 29 percent. HPC 3. Foxtail pine (P. aristata). 24 inches DBH. Midway between HPC 1 and 2. Average width of growth layers 1.32 mm. Range 1770-1946. Growth- layer sequence variable. Average departure from mean 0.51 mm., or 39 percent. HPC 4. White fir (Abies concolor). 15 inches DBH. Wettest location for group east of Pass; slight drainage toward tree. Average thickness of growth layers 2.93 mm., corrected to average 1.46 mm. Range 1880-1946. Growth-layer sequence only fairly variable. Average departure from mean 0.57 mm., or 39 percent. HPC 5. Douglas fir (Pseudotsuga taxifolia). 15 inches DBH. Average width of growth layers 1.16 mm. Range 1810-1946. Growth-layer sequence variable. Average departure from mean 0.45 mm., or 39 percent. HPC 6. Foxtail pine. 14 inches DBH. Distant 20 yards from HPC 5. Aver- age width of growth layers 1.38 mm. Range 1820-1946. Growth-layer sequence uniform. Average departure from mean 0.36 mm., or 26 percent. HPC 7. Ponderosa pine. 23 inches DBH. Distant 70 feet from edge of terrace above stream. Average thickness of growth layers 1.91 mm. Range 1830-1946. Growth-layer sequence uniform and rhythmic. Average departure from mean 0.52 mm., or 27 percent. HPC &. Ponderosa pine. 13 inches DBH. Distant 20 feet from edge of terrace above stream. Least mature. Ground-water relations make site better drained than that of HPC 7. Average thickness of growth layers 3.74 mm.; corrected to average 1.46 mm. Range 1897-1946. Growth-layer sequence only fairly vari- able. Average departure from mean 0.53 mm., or 36 percent. HPC 9. Ponderosa pine. 19 inches DBH. Midway between terrace edge and base of steep slope. Wettest location for trees west of Pass. Average thickness of growth layers 2.71 mm.; corrected to average 1.46 mm. Range 1857-1046. Growth-layer sequence variable. Average departure from mean 0.44 mm., or 30 percent. The designations variable, fairly variable, or uniform were assigned directly from the wood by visual judgment alone. No. 18 TREE GROWTH AND RAINFALL—GLOCK 5 METHODS Because all cores consisted of sound wood, none was discarded. Furthermore, because site factors such as light, drainage, slope, ground-water relations, and competition were evaluated on the spot as closely as possible, no reason existed immediately after the col- lection had been made for the rejection of any specimen. The collec- tion was considered a normal representation of the site factors at the three chosen localities even though the sequences differed to a great extent in variability and average growth-layer thicknesses. At the time the cores were taken there seemed to be no reason why different species should show differences except those due to slight variations of site factors peculiar to each tree. Such a factor as soil aeration had to be judged by soil texture and composition and visible soil-water relations. There was no opportunity for analyses or measurements. Indeed, this problem of selection in the field, with- out measurements, was of great importance: could local site factors be judged with sufficient accuracy to demand the inclusion of each core as a representative specimen in the general collection? If so, choice in the field, based on ecologic principles, would be a dependable method of selection whose integrity could be questioned only on field evidence or its derivatives. Treatment of the wood.—The cores and the growth layers they contained were subjected to the following procedure to prepare them for correlation among themselves and with rainfall. 1. The cores were glued in a groove sunk into the curved side of half-inch half-round and “shaved” by razor sufficiently to expose the growth layers clearly.° 2. Beginning with the increment for 1946, which was complete be- cause of the time of sampling, October 5, 1946, the growth layers were counted inward and dated on the assumption that each sharply bounded layer represented a year. 3. Skeleton ® plots were set up on coordinate paper, each ordinate representing a year. If a sharply bounded growth layer was decidedly thinner than its immediate neighbors an ink line was drawn on the ordinate appropriate to its date, the height of the line being inversely proportional to the thickness of the growth layer. The resultant skeleton plots and the master plot derived from them are shown on 5 Principles and methods of tree-ring analysis. Carnegie Institution of Wash- ington Publ. No. 486, p. 6, 1937. 6 Thid., pp. 14-16. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III figure 2. Thus, the wood specimens were cross-dated with one another ; that is, growth layers taken to be equivalent in time were set in line with one another. 4. The thicknesses of the growth layers were measured to hun- dredths of a millimeter by means of a measuring microscope. These measurements are called raw data in millimeters. 5. The average thicknesses of the growth layers on the sequences HPC 4, 8, and 9 were corrected downward to approximate the averages of the other sequences. Otherwise, if a sequence of high average thickness were one of several merged into a group, its high average would unduly influence the values in the group. 6. The raw data in millimeters of each sequence were changed into percentages of the sequence mean in order to establish an identity of units and an identity of base line between tree growth and rainfall. 7. The raw percentages were smoothed by the formula a+2b+c 4 8. Various sequences were merged into groups and smoothed. Groups.—The nine sequences divided themselves geographically into three groups—east of the Pass, on the Pass, and west of the Pass. Nevertheless, other groupings were arranged in order to make the tests not only as critical but also as thorough as possible so far as comparison with rainfall was concerned. The following groups were set up: Group 1 (G1). Trees 1-5, 8, and 9. Variable and fairly variable sequences only. Group 2 (G 2). Trees 1-3, 5, and 9. Variable sequences only. Group 3 (G 3). Trees 1-3, and 9. Most variable sequences based upon a visual study of the wood samples. Douglas fir omitted. Group 4 (G 4). Trees 1-4. East of the Pass. Group 5 (G5). Trees 5 and 6. On the Pass. Group 6 (G6). Trees 8 and 9. West of the Pass, exclusive of the tree whose sequence is uniform. Group 7 (G7). Trees 1-9. All trees. Group 8 (G 8). Trees 1, 2, and 9. Ponderosa pines with variable sequences. Group 9 (G 9). Trees 7-9. West of Pass. Group 10 (G 10). Trees 1-3, and 7. From the drier sites. Group 11 (G11). Trees 4-6, and 9. From the wetter sites. The primary groups are numbers 4, 5, 9, 10, II, and 7. Selection of rainfall stations and the treatment of data—A moun- tainous country permits little choice in the selection of rainfall stations. Fortunately, one station, Chacon, lies approximately 7 miles, airline pe a *29.1} 94} JO JOAe] YJMOIS [eI} -99 dy} UO $}Sd1 d]IIID & JO apisul jop Y ‘s1oAe] YJMOIS MOI1eU JO dnois & sapnyour auly AABM YW N ‘yory} suvoU , T,, pue a[qQnop suvaw .TgqC,, “JoAE] YIMOIT oY} JOMOIIEU JY} ‘SI BUTT PAYUI oy} JoSuo] BY “}0[d s0jseur & O}UT stsoyjUAs 1194} pue “6 03 I DGH ‘S901} outu 9y} Jo sjojd uojsjayG—ze “SIA O+6l Of6! O26) Ol6l OO6l O88) O88 OLB OF8l OSI OFS! OFI | WILSV) | | 3 aNIid 1 () we o y J oe: Tt 4 3aNIid | 4 rr aoe Z 4 It La iy ad q P| aNId fy = < or ‘a Z < ee H = S) 4 oO fy 1¢3) 4 H i) ees eal — S Z Pal cc < Wy a} a 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III distance, north-northeast of Holman Pass. Unfortunately, its record is short compared with that of Santa Fe, distant 4o miles to the southwest from the Pass. Figure 1 shows the general relations. Pertinent data in regard to the stations follow: Chacon: 7 miles north-northeast, east of divide, in the mountains. Elevation 8,510 feet. Length of continuous record 1909-1941. Black Lake: 19 miles north-northeast, east of the divide, in the mountains. Elevation 8,348 feet. Length of continuous record 1909-1946. Taos: 24 miles north-northwest, west of mountains. Elevation 6,983 feet. Length of continuous record 1901-10945. Taos Canyon: 32 miles north, west of divide, in the mountains. Elevation 8,059 feet. Length of continuous record 1909-1941. Las Vegas: 35.5 miles southeast by south, east of mountains. Elevation 6,400 feet. Length of continuous record 1887-1943. Santa Fe: 40 miles southwest, west of mountains. Elevation 7,013 feet. Length of continuous record 1850-1944. Albuquerque: 1o0t miles southwest, west of Sandia Mountains. Elevation 5,196 feet. Length of continuous record 18092-1946; partial record 1850-1861, 1863-1867, 1878-1879, and 1889-1890. Month-intervals chosen for the correlative tests between rainfall and tree growth were: November-May May-July January-May May-August January-August April March-April May March-June June March-July July May-June August Rainfall data were then subjected to the treatment here outlined: 1. Addition of monthly rainfall totals in order to obtain the rain- fall of the intervals listed above. These sums gave raw data in inches. 2. The raw data of each interval in inches were changed into per- centages of its own mean in order to establish an identity of units and an identity of base line between rainfall and tree growth. 3. The raw percentages were smoothed by the formula a+2b+c 4 4. The raw percentages for the stations Chacon, Black Lake, Las Vegas, and Taos Canyon were merged into a group record for the intervals January-August and March-July. These stations were chosen because their interval averages were nearly the same. Method of correlation Because the purpose of the study was the correlation of rainfall variation with growth variations and because No. 18 TREE GROWTH AND RAINFALL—GLOCK 9 the basic data constitute a continuous time series, the trend method 7 of correlation has been used. This method, in the coefficient t, gives a measure of parallel variation combined with amount of that varia- tion. If ¢ equals 1.00, the trends in both sets of data, tree growth and rainfall, are wholly parallel or in the same direction; if t equals —1.00, the trends are wholly opposite. Tests were constantly made to detect the undue influence of one or two entries should such be present. The ratios of opposite to parallel trends are included in the tables with the trend coefficients. With the trend method of cor- relation, secular trends or long-period fluctuations did not have to be eliminated. These are ignored for the present especially because the history of the stand and the histories of the individual trees are unknown except as revealed on the wood itself. The quality of the correlations between tree growth and rainfall on identical years was tested by application of a one-year lag and by reversal of data. In all cases, the correlations dropped to a value of no significance. Although the purpose of the work was the comparison of year-to- year variations, correlations involving smoothed data (second inter- mediate) were nevertheless carried through the main part of the calculations. The majority of the coefficients did not increase signifi- cantly over those using raw data; in fact, many decreased. The initial questions, then, to be answered by use of the trend method, were: If rainfall increases or decreases, does tree growth, as shown by such simply obtained samples as increment cores, increase or decrease in like direction? To which rainfall interval does the tree growth correspond? As the work progressed new problems came to light and soon carried the study far beyond the original objectives. STUDY OF THE GROWTH LAYERS Cross-dating —This process consists of establishing the identity in time of growth layers on different sequences by matching narrow growth layers, in particular, from one specimen to another. Judg- ment as to narrowness depends upon visual comparison with im- mediately adjacent growth layers and should not be confused with or substituted for mathematical expressions. Obviously, cross-dating can possess various degrees of excellence. What constitutes reliable cross-dating is a moot point and may, perhaps, be largely dependent 7 A rapid method of correlation for continuous time series. Amer. Journ. Sci., vol. 240, pp. 437-442, 1042. IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III upon the individual investigator. Here, it is important to know if cross-dating is a prerequisite to the merging of sequences for correla- tion purposes. Figure 2 shows the so-called skeleton plots for HPC 1 to 9 and a master plot made by a synthesis of the nine. The heights of the inked lines bear an inverse ratio to the widths of the growth layers on the wood as judged by the eye. No actual measurements enter the skeleton plots. In order to judge the quality of the relationships, the above figure should be compared with figure 3, which shows excellent cross-dating from a forest-border area. The conclusion is obvious: cross-dating as exhibited by the Holman Pass specimens is of remarkably poor quality. One is tempted to say it does not exist at all, for, if the dates were entirely unknown and within a range of several centuries, one would have difficulty in convincing others that the sequences match growth layer for growth layer as they stand. In the present case the validity of the cross-dating, or the only assurance that the growth layers grew on the dates assigned to them on the skeleton plots, rests on two circumstances: (1) the narrowness of the growth layers designated 1880 and 1893 and (2) the probability that the soil moisture in the zone where the trees grew was sufhciently adequate to prevent a temporary halt in growth during any one growing season. A detailed comparison of the growth layers on all specimens for each date in succession (fig. 2) brings out a lack of correspondence that appears to emphasize a certain degree of individuality in the site factors at each tree. An analysis of figure 2 was made, and the results were arranged in table 1. The record covers 116 years. Out of this length of record only two cases exist, the growth layers designated 1880 and 1893, wherein the growth layers are notably narrow on the eight specimens bearing them. Two cases, 1836 and 1934, could perhaps be classed with the previous two because the one growth layer that does not conform on each exists on a so-called uniform sequence, one on HPC 7 and the other on HPC 6. Entries in table 1 wherein no narrow growth layer exists on any one of the specimens number 55 cases. With the number of specimens in the count disregarded, there are 22 cases where one growth layer is atypical, 23 cases where two growth layers are atypical, 12 where three are atypical, and 2 where four are atypical. For more than half the years the sequences are from II to 50 percent out of agreement with one another. TE ‘z oinSy uo uMOYS SUl}ep-ssot9 1ood oy} YIM pasedwios oq pynoys AoyL “BUOZNY UIBYIIOU UT JopsJog }sotoJ 9y} 1% MaIS yey} Seo1} Jesaides ssoy, “SUNeP-SSO4) yuaqjaoxa Surmoys sjojd woysyo>s peysyeyw—e “oI 09 008! 08 09 OF O22) TREE GROWTH AND RAINFALL—GLOCK II! VOL. MISCELLANEOUS COLLECTIONS SMITHSONIAN TABLE 1.—Incidence of growth-layer type Analysis of figure 2 4a rr “AY MOL1E NY 14a STL “AY MOLIEN Buta AOL “AY MOLIV NT --0 LOT -O ins keAlecie c WUNAWD MOAN ONO OO TMO NOMNNRWWNO O owt Or ONO: 16 al ine] ooow nwmmw No oO O eo 1921I.. N Pare § TOOWer WwW ite. they AA NNANANWO AN NN ONONMON ODADANHDHDAMAM (Wj toy fo} {oe Tab (2) {ey fe) {9} ~+HNuUNUMNwO wo H NH He HH OO OO oo al TOZT sicte 6 ™~ iol N + ANH ATOAOO NN ONDOoOMnA HAN N CmMOooMN OO NMO ¥FMONMOON TW Peo VL X= BCT =e ce iy | .0 TOAD.) <2 (e/a) 0 TQ OiLietels TS6L. «2 omOoNn to maoanwowd nw nwo MonNnN OMOm ray is: Nn NN + oo coc om NoveEs: 1836—an average growth layer on HPC 7, a uniform sequence. 1934——an average growth layer on HPC 6, a uniform sequence. No. 18 TREE GROWTH AND RAINFALL—GLOCK 13 Correlation.—Because of the poor quality of the cross-dating and because of the desire to compare the sequences each with the other based upon precise measurements, the sequences were subjected to statistical correlation. Table 2 gives the trend coefficients and the ratios of opposed to parallel trends for certain trees and certain groups. The bases of selection are evident from the captions in the table. A comparison of the averages for uniform and variable sequences shows that the uniform have a considerably higher trend coefficient, TABLE 2.—Holman Pass collection Trend coefficients and ratios of opposed trends 1850 1898 1850 1898 to to to to 1897 1941 1897 1941 Uniform sequences Trees east of Pass 4 vs. 6 0.82 (0.36) I vs.2 0.62 (0.40) 0.90 (0.23) 4 VS. 7 0.83 (0.41) I vs. 3 0.86 (0.34) 0.94 (0.20) 6vs.7 0.66 (0.38) 0.57 (0.36) I vs. 4 0.79 (0.32) 4 vs. 8 0.88 (0.27) 2 vs. 3 0.85 (030) 0.97 (0.11) 2 VS. 4 0.83 (0.23) Variable sequences 3 vs. 4 0.96 (0.25) Tvse5| 0.36 (0.53) 0:73 (0.20) Evs.9 0.32 (0.42) o.81r (0.23) fevsa9 020 (0.42) 0.71 (0.25) 5 vs.6 0.79 (0.34) 0.91 (0.20) Trees on Pass Intergroup Trees west of Pass 4 vs.5 0.90 (0.32) 0.88 (0.20) 7 vs. 8 0.87 (0.23) A-YWS39) 0:72 (0.36) 0.87 (0.23) 7 vs.9 0.72 (0.40) 0.63 (0.30) 5 vs.9 0.52 (0.49) 0.76 (0.27) 8 vs. 9 0.70 (0.30) IO vs. II 0.94 (0.21) 0.97 (0.14) whereas the variable have a very slightly lower trend ratio. If any- thing, the uniform have a slight advantage. Intragroup comparisons east of the Pass, on the Pass, and west of the Pass indicate on the whole that correlations are distinctly higher between trees within their own groups than between trees in different groups. If the groups are averaged, this higher correlation is shown even more clearly. Hence, trees grouped together appear to correlate more closely than those rather widely separated. The matter of distance merits further attention. If the trend ratios of the trees in table 2 are arranged in order of distance within the T4. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III group, table 3 results. Among the species, PP means ponderosa pine, FP foxtail pine, WF white fir, and DF Douglas fir. The interval of years in either case ends on 1941. Table 3 is divided into three groups: the first comprising trees east of the Pass, the second on the Pass, and the third west of the Pass. In general, agreement declines with increasing distance, a distance measured in feet. Site factors at the surface appear to the eye to be nearly identical among the trees of any one group, but apparently the factors do change within short distances in spite of appearances. Proximity outweighs difference of species as well as presence or absence of variability. Factors present at the immediate location of the individual tree, or what may be called microsite factors, appear TaBLe 3.—Holman Pass collection Ratios of opposed trends Distance Total apart Trees Species 1898-1941 sequence 150 feet ZN Sons PP vse I vs. 3 PP vss EP 0.18 0.24 2 vs. 4 PP vs. WE 300 D vse 2 RIP we, IRIE 0.24 0.29 3 VS. 4 FP vs. WE 450 I vs. 4 PPiyse Wir 0.32 0.34 60 (ca.) 5 vs. 6 DF vs. FP 0.20 0.27 50 7 vs. 8 PPSvsa2 0.23 230 TEN SHeO IBIS sys, IEIE 0.30 8 vs. 9 IPI) yy, IRN to exert a strong measure of control on tree growth. That trees separated by a distance of a mile or more do show a parallel agree- ment of variation in a majority of years indicates the influence of a gross factor uniformly variable within limits over the area. How- ever, when it is remembered that the trend between two growth layers on one tree compared with the trend between two growth layers of the same date on another tree, no matter how remote, can vary only in two directions, parallel or opposite, some allowance must be made for accidental similarities. The same principle, of course, holds true where visual comparisons are made in so-called cross-dating because in the consideration of two growth layers of same date in different trees one growth layer can only be thinner than, thicker than, or of the same thickness as, the other growth layer. no. 18 TREE GROWTH AND RAINFALL—GLOCK IS Correlations between groups (table 2) east of Pass, on Pass, and west of Pass are only fair. They show a mixed influence of site and distance. Groups 4 and 5 are relatively close together but have dis- similar sites—they have the highest correlation ; groups 4 and g are far apart but have somewhat similar sites—they have correlation of intermediate value; and groups 5 and 9 are far apart and have very dissimilar sites—they have the lowest correlation. Table 2 suggests something much more surprising than the dominant influence of local site factors. The correlation among different trees and among different groups as shown not only by the trend coeffi- cients but also by the trend ratios are distinctly less for the period 1850-1897 than for the period 1898-1941. In fact, a few of the trends, and trend ratios, are of such poor quality as to indicate little relationship. Growth factors from 1850-1897 apparently must have had a localized variability which to a certain extent became less localized after 1897. For further comparisons among the trees the trends were plotted for each tree against every other tree for the total years of record. A comparison of ponderosa pine with other ponderosa pines, of ponderosa with other species, and of other species among themselves shows that species has no bearing upon the trend agreements. A comparison of sequence types, such as variable with variable, variable with uniform, and the like, shows that the type of sequence being correlated is not an important factor. In general terms, however, the closer two trees are together the greater the number of parallel trends. During the period of 44 years from 1898 to 1941, where all nine trees are in the record, there are 15 years with parallel trends. Agreements are concentrated in the 10-year period, 1920-1929, which has 6 parallel trends. Back of 1898, the period of 48 years adds only 9 parallel trends to the 15 of the later period in spite of the fact that the record of the earlier period contains from one to three fewer trees. The striking lack of agreement prior to 1898 appears to fit in with the lack of correlation mentioned in the paragraph above. Again it seems that the microsite factors may have contrasted more acutely from tree to tree or that an over-all factor exerting a general influence on tree growth may have been more areally variable than later. 10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Lit Growth-layer characteristics—Table 4 shows the average of year- to-year variations of growth-layer thicknesses on single tree sequences and on three groups. These figures are the measured equivalents of the visual values embodied in the terms variable, fairly variable, and uniform. In the main, the numerical results militate against judgment by eye. Tree HPC 5, for instance, was judged variable and HPC 6 uniform ; yet both have nearly the same average variation. However, greater consistency is shown by groupings: for the period, 1898-1941, the average of the variable sequences is 0.36, of the fairly variable 0.32, and of the uniform 0.28. Table 4 emphasizes the importance of location, not species, as the apparent determinant of average variation. For instance, trees HPC 3 and 6 are both foxtail pines and yet have variations of 0.37 and TABLE 4.—Holman Pass collection Average year-to-year variation 1898 1850 1898 to to to to 1897 1941 Entire 1897 1941 15 le Gat nou adie 1850- 0.22 0.40 0.30 GoTONco eae 0.26 0.40 BS aon 1850— 0.35 0.41 0.38 Teli eete serene 0.38 0.27 Biaciee oisens 1850-031 0.43 0.37 SEE Ta 0.21 0.27 Aes teeerhees 1880- 0.36 0.28 0.30 Si oR Alaic 1850— 0.29 0.19 0.25 (\ ARR eos 1850- 0.27 0.21 0.24 Fine stated 1850-018 0.35 0.26 Oe ciecieters 1857- 0.40 0.39 0.40 0.24, respectively (0.43 and 0.21 for 1898-1941), the higher value existing in the drier location. It is true that the ponderosa pines have higher variations in general than the other species but HPC 3, a fox- tail pine, grew between HPC 1 and 2 and has even a slightly higher average variation. All trees from the wetter locations (group 11) had higher average variations for the period 1850-1897 than they did for the period 1898-1941. In contrast, the trees in the drier locations (group 10) had lower average variations in the earlier period. Tables 5 and 6 giving average growth-layer thicknesses and average departures were prepared even though definitive results were not expected because secular trend and long-period fluctuations had not been eliminated. In table 5, group 10 shows an increase and group II a decrease of average growth-layer thicknesses from the period 1850- No. 18 TREE GROWTH AND RAINFALL—GLOCK 1 7/ 1897 to that of 1898-1941. The individual trees of group Io are not consistent among themselves in that HPC 1 and 3 increase decidedly, HPC 2 increases very slightly, and HPC 7 decreases. All trees in group II are consistent except for HPC 6 which decreases very slightly. Thus, four trees decrease, two remain practically unchanged, and two increase their average thicknesses for the period 1897-1941 contrasted with that of 1850-1897. Group 7 reflects these influences. TaBLE 5—Holman Pass collection Average growth-layer thicknesses — 18968 1850 1898 to to to to 1897 IQ4I 1897 1941 EPC ees oss 1850— 1.28 1.82 Gita 2 1.53 17 2 tne 1850- 1.87 1.88 1 Corel 2.26 1.96 Cl eee 1850— 1.01 128 eae ere 1.90 1.83 Aaa ha 1880- 3.50 2.95 arcuate 1850—- 1.18 0.79 Octneaee 1850- Le52 TS 7 Ona 1850- 1.96 1.86 One eet. 1861- 2.85 2.61 TasLe 6.—Holman Pass collection Average departures 1898 1850 1898 to to to to 1897 1941 1897 1941 Te GT vee soci) 5 1850- 0.31 0.36 GalOeseecer 0.32 0.36 7) eee 1850- 0.32 0.27 Die ese coke 0.34 0.30 Sun neers 1850— 0.45 0.42 OT too Sats ie 0.18 0.20 Ab ee 1880- 0.37 0.36 Ciaer rots + 1850- 0.39 0.34 Oyo eee 1850— 0.27 0.21 ce 1850- 0.19 0.38 openers hs: 1861- 0.31 0.28 In Table 6, group 1o shows an increase and group 11 a decrease of average departures from the period 1850-1897 to that of 1898-1941. All trees of group II are consistent among themselves in the decrease from the earlier to the later period. This is not true for the trees of group 10. Two of them, HPC 2 and 3, actually showed a decrease of average departures and thus conformed with the wet-site trees of group 11. In other words, trees HPC 1 and 7 do not conform with the remaining six trees, yet their influence is sufficiently great to determine the relative values as shown for groups 7 and 10 in table 6. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III The data in table 7 were calculated in an attempt to obtain a measure of excess variation over normal. From the earliest to the latest periods shown, HPC 1, 2, 3, and 4 show a rise and decline; HPC 5, 6, and 9 show a general decline ; and HPC 7 shows a general rise of values. Six of the trees, but not including HPC 1, have lower values for 1910-1941 than for 1850-1897. As in the case of average departures, it is HPC 1 and 7 which do not conform, In spite of their influence, group 7 shows a slight but progressive decline from the earliest to the latest period. If HPC 1 and 7 are eliminated from group 7, giving group 7 (restricted), the decline becomes more decided. TABLE 7.—Holman Pass collection Average departure from mean variation 1850-1897 1898-1941 1910-1941 FDP Cit ia che ae aoe 0.17 0.28 0.24 Dora dic Sra ard MMI TE He 0.21 0.24 0.20 Daeg slag eto oe 0.25 0.26 0.23 Avutsscusc mee eee eee 0.23 0.24 0.21 Shauaiaioleicesenaeeiaroveia apes iat 0.20 0.13 0.11 Ghana ch Eee 0.19 0.16 0.15 Tote MEIC OSE hee 0.15 0.23 0.24 Once ec eee 0.28 0.25 0.20 Kr TOR Aatacey stat titel aeaeuenstnre 0.132 0.148 0.136 Tela tanayswaratons: caevers Chemie ones 0.204 0.118 0.110 MRS aA BOO BSD Gabino 0.110 0.105 0.004 7.0 (restricted) yer. ia. 0.130 0.105 0.092 Table 8 brings together a short summary of characteristics on the wood in order to emphasize the differences between the two periods 1850-1897 and 1898-1941. Although the differences between groups Io and II appear striking, they actually are due to the influence of two out of eight trees. Elimination of those two trees from group 7 brings it into harmony with group 11. There remain, then, the fundamental differences between the periods 1850-1897 and 1898- 1941. Do they reflect a change in amount of rainfall with its attendant changes in rainfall characteristics, or a change in the rainfall interval important to tree growth, or both, or some other change? In a pre- vious paragraph a striking dearth of trend agreements among the trees was pointed out for 1850-1897 in contrast with succeeding years. A reexamination of the data shows that the dearth does not apply quite so drastically to the trees from the wetter locations. This No. 18 TREE GROWTH AND RAINFALL—GLOCK 19 matter of trend agreement appears to be another facet of the general problem brought out by the changes of characteristics on the wood through the years from 1850 to 1941. TABLE 8.—Holman Pass collection Characteristics 1850-1897 1898-1941 Average variation Ee NOS oe epnee Oto le esiehans 0.26 0.40 10 cS acicuOreeen Pitot eee tee 0.38 0.27 Tisealie rai stone slick stays Fae avalon 36 0.21 0.27 GTO a beta ree ok Anes aah 1.53 1.70 TGS ead AGN acne a6 1.96 Ti Ses oe he 1.90 1.83 GRBLOM eR ners Seats sols: 0.32 0.36 Ely sates eyed anise eae 0.34 0.30 The beta Rt od Ce Cee Gee 0.18 0.20 (GEO rete ee eens cee eke 0.132 0.148 1 oan aes ei RR aren ae ED 0.204 0.118 FAIR RERTR eS ol es 8 0.110 0.105 STUDY OF RAINFALL CHARACTERISTICS Two tasks were set out for consideration in connection with the rainfall data: (1) to determine the interrelationships among the stations of usable records in the vicinity of Holman Pass and (2) to determine the characteristics of those records. As to the first task, it is necessary to know the extent of the differences between two adjacent stations in order to appreciate and allow for the possible differences between the trees and the station nearest to them. As to the second task, it is desired to learn whether or not the rainfall shows any differences between the two periods 1850-1897 and 1898-1941, and, if it does, to compare the differences with those obtained from a study of growth-layer sequences. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III Interstation correlations—Table 9 shows the trend coefficients and ratios of opposed trends between Chacon, the nearest station to Holman Pass, and six other stations for eight selected time intervals. These intervals were chosen on the basis of their possible influence on tree growth. On the whole, the correlations show a remarkable con- sistency. Those comparisons which do not include part or all of the summer rainfall are commonly higher than those which do. Furthermore, the longer the interval under comparison is, the poorer the correlation in general. Black Lake, the nearest to Chacon in distance as well as elevation, does not have the best correlation with Chacon. Las Vegas has the greatest similarity, a station farther away, 2,100 feet lower, and out beyond the foot of the main range of mountains. Santa Fe rainfall correlates with that of Chacon to a degree equal to the correlation between Black Lake and Chacon. Even Albuquerque is little less in degree of similarity. The best correlations are for the March-April intervals with Black Lake and Albuquerque which show ratios of opposite trends with respect to Chacon of 0.12 and 0.09. It is scarcely necessary here to do more than refer briefly to the many observations of differences in rainfall at gauges spaced rather closely together. For instance, Stout § records a study of July 1948 rainfall on a plot centering at El Paso, Ill. Two stations, 10 miles apart, had 10.44 and 5.93 inches of rainfall. Two other stations, 3 miles apart, showed a difference of 77 percent. Localization of single storms is on occasion even more pronounced. On June 30, 1947, near Lubbock, Tex., 4 to 5 inches of rain fell in a belt about 2 miles wide, whereas none fell 2 miles to the west and 0.26 inches 8 miles to the east. Of course, this may be unusual, but at least it is more or less typical of extreme forest-border conditions. Furthermore, it must be remembered in comparing tree growth with the rainfall of a station that, as pointed out by Landsberg,® a rain gauge samples but does not measure rainfall and therefore “the areal significance of precipitation amounts caught at a station is very re- stricted. . . .” These characteristics of rainfall must be duly weighed when the growth of selected trees is compared with the record of a station some miles distant. The trees may respond to the rainfall they themselves receive but differ somewhat from that received by the weather station. 8 Weatherwise, vol. 1, pp. 112-113, 1948. ® Landsberg, H., Critique of certain climatological procedures, Bull. Amer. Meteor. Soc., vol. 28, pp. 187-191, 1947. TREE GROWTH AND RAINFALL—GLOCK No. 18 (910) coo (60°0) g6'0 (€z'0) 060 (910) £0'0 (910) £6'0 (SZ'°0) 00'0 (Z1'0) 66'0 idyw ~ IRIN (610) 88°0 (920) 940 (£70) 83°0 (€1'0) £6°0 (610) v60 (£z'0) 93°0 (610) z6'0 ACI “AON SUOLIDIS (610) (gI'0) (Z£"0) Q9"0 060 So (910) (Sz’0) (170) 93°0 z6°0 990 (Zz'0) (Z1'0) (VE'0) 930 S6'0 93'0 (91°0) (610) (82°0) S6'0 £30 16°0 (61'0) (910) (4£°0) 16°0 z6'0 93°0 (1£°0) (610) (Zz'0) 09'0 cgo £30 (61'0) (91°0) (82°0) 880 z6'0 £6'0 ACW oun “any uel -ABTY -APTAL IPO6I-6061 (120) +6'0 (2Z'0) F6'0 (Z1'0) 360 (Zz'0) £6°0 (Zz'0) S6'0 (Zz'0) 63°0 (gz'0) £0°0 ounf SPaai\g (¥z'0) ¥g'0 (1£°0) 9370 (1£°0) AO) (Z1'0) 10'0 (82°0) 93°90 (Zz'0) 920 (ZZ'0) £6°0 ‘any uel spuasy pasoggo fo soups pun syuaiif{a0a puas J stot OSRIDAY 6go ‘'t'***** anbsonbngqyy 4ayj0 fo ywyy pun jjofuws uosDYyD Uaanjaq Uo1jwjas4oJ—6 HAV L o89* 9g BURG "** SBSIA SBT uoAuey) soey, ats heel" SAP NTE "+ aye] pe i) bo SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. fit From the qualitative standpoint, the trend ratios of table 9 give a rather clear indication of the amount of agreement to be expected between tree growth and rainfall where the two are as far apart as any two of the rainfall stations. Quantitatively, trend coefficients yield values to be expected in the same fashion. If variations in tree growth mirror variations in rainfall to a high degree then the cor- TABLE 10.—Correlation between rainfall intervals at Chacon Trend coefficients and ratios of opposed trends 1909-1941 Novy.-May vs. Jan.-May vs. JamssAnagt Mescptieschenes 0.78 (0.26) Mari-A pres sence 0.92 (0.22) Nov.-May vs. Jan.-May vs. Fani=Miayin-caeee. cure 0.99 (0.06) Maris}tineleee. cee 0.84 (0.22) Novy.-May vs. Jan.-May vs. Martine eee 0.82 (0.29) May-Aug). scae. eee 0.14 (0.47) Jan.-Aug. vs. Jan.-May vs. Jans-Mayes Src see 0.73 (0.28) May—Jranen eee. set 0.18 (0.44) Noy.-May vs. Mar.-July vs. Mar —luly semen 0.77 (0.23) ManroApr nese 0.67 (0.31) Nov.-May vs. Mar.-July vs. Vikiy Ania nies at cle 0.21 (0.48) Mar -Juneé (2.24.2 0.08 (0.12) Nov.-May vs. Mar.-July vs. Wiehvealheine Bho a5 000 0.17 (0.45) May-Atigcne ae saeaee 0.85 (0.34) Nov.-May vs. Mar.-July vs. Mar sA prs 4 aesteotan 0.90 (0.23) May=Junel a. ec ae 0.81 (0.22) Jan.-Aug. vs. May-Aug. vs. Mar June iano asec 0.87 (0.25) May-June secre 0.04 (0.25) Jan.-May vs. May-Aug. vs. Mar —Jitly; sence 0.72 (0.22) Mar-Apr Gexoues —0.33 (0.66) Jan.-Aug. vs. May-Aug. vs. Mart=Iiuilyg nee 0.97 (0.22) Mari= June t..6c0ehs 0.66 (0.44) Jan.-Aug. vs. May-June vs. May-Atge. 22. ....-- 0.92 (0:25) § Mar-Aprisse. ues —0.02 (0.47) Jan.-Aug. vs. May-June vs. May=Juney ase es: 0.69 (0.22) Mate ay une) janet 0.87 (0.25) Jan.-Aug. vs. Mar.-Apr. vs. Mar Appaemaser tice 0.56 (0.38) Marisunes: cee secee 0.90 (0.25) relation between Holman Pass trees and Chacon rainfall should closely approach or possibly equal the average values set out in table 9. Should this prove to be true, the conclusion is no doubt justified that trees growing in a zone well above critical moisture conditions rather faithfully record rainfall variations at the site from year to year, barring the impact of an “accidental” factor in concentrated form in any one year. Table 10 sets forth correlations between various month-intervals in ee No. 18 TREE GROWTH AND RAINFALL—GLOCK 23 Chacon rainfall. Good correlations in general result under three circumstances: In the lack of summer rainfall in the intervals com- pared, in the proportion of overlap between the two intervals, and in the length of the intervals. For instance, November-May versus January-May has a trend coefficient of 0.99 and a trend ratio of 0.06 whereas, contrariwise, May-August versus March-April has values of —o0.33 and 0.66. The table as a whole shows great variation and indicates the necessity of comparing tree growth with different rain- fall intervals. It goes farther than this. If tree growth is found to correlate with one particular month-interval, then a great quantity of xylem (as a thick growth layer) formed during a certain season suggests copious rainfall for that month-interval; it does not neces- sarily suggest that the entire year is a wet one. Rainfall characteristics——For a study of the influence of a single factor, such as rainfall, on tree growth it is necessary to have long records at the immediate site of the trees. Short records taken a matter of several miles distant can be highly indicative but not neces- sarily conclusive. In the present case the record at Chacon, 7 miles away, begins with 1909. Therefore, the longer records of Santa Fe and Albuquerque were used, in spite of greater distances, in order to determine possible differences in rainfall characteristics between the periods 1850-1897 and 1898-1941. Table 11, in the first place, gives the March-July and January- August rainfall for 1909-1941 at the several stations. As will be shown later, the rainfall of March-July is a significant factor in tree growth. In the second place, table 11 gives the rainfall of Santa Fe and Albuquerque for the periods 1850-1897 and 1898-1941 set out for various month-intervals. Two points must be considered. First, there is the striking fact that the average rainfall of March-July for both Santa Fe and Albuquerque was less during the period 1850-1897 than during that of 1898-1941. The same is true for the average rainfall of Albuquerque for January-August. Second, there is the fact that the average rainfall of January-August at Santa Fe was greater during 1850-1897 than for the following 44 years. The reason for this inconsistency with the intervals mentioned in the first point above was suspected as soon as it was determined that the average rainfall of January-May, in contrast to the rainfall of January-August, was less during the earlier period, 1850-1897. Therefore, the average rainfall was computed for March-April, May-June, April, May, June, July, August, and September. Only July and August showed greater 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOE. i average rainfall for 1850-1897 than for 1898-1941. It was obvious at once that the greater rainfall of August aided by that of July caused the greater average rainfall of January-August during 1850-1897 at Santa Fe. In September, as a matter of interest, the averages swing back so that the figures are 1.58 inches for 1850-1897 and 1.66 inches TABLE 11.—Average rainfall (inches) March-July January-August 1909-1941 1909-1941 Chaconeatian: scrcsrenstaciecarrseces 10.34 15.80 Black Walere cena cece ele 8.42 12.66 PAGS efi nadine a apae ests. 5.96 8.83 diaosy Ganvonge eee oe eee 9.64 14.68 as, Ve ass deaths esichos aie: 9.33 13.66 Santa. Hes scmreamivtnceheiesersienieaors 6.88 10.14 Al buquenquelmerce enor: 4.2 6.28 Santa Fe Albuquerque 1850- eat 1850- 1808- 1897 1941 1897 1941 IMarcholiuly, teaser ieee 6.34 6.74 3.10 3.07 ETE SA CANIeAbSIEy cacecgagceoe 10.60 10.11 5.60 5.92 Janay Way eae o 3.04 4.82 1.75 2.50 MarchoAprill tnccis eeraeisase 1.54 1.92 0.58 1.07 May=Junerts oc Ja ihactande cies 2.14 2.61 1.26 1.43 April ee ane Bie ees cela 0.71 1.08 0.29 0.67 IMB: Fides onto ars drakvantelne teks 0.99 1.19 0.36 0.73 ASIEN aioe Sevsc beaks elaeyy Re ep hese 1.15 1.18 0.79 0.69 A UAL: Aivags cis euskorcvea teacher Reeve ee 2.66 2.21 17, 1.47 AUIOUSt. Nera ntiednlc ee cc pigtier ves 2.89 1.90 1.59 1.26 Septembery deccisette sine erence 1.58 1.66 0.90 0.08 Albuquerque, 1850-1897, 27 years of record only. for 1898-1941. This situation no doubt should be given emphasis: the average rainfall for the interval March-June began to increase somewhere near 1898 whereas that for the interval July-August decreased. Calculation shows that the average for March-June began to increase slowly just before the turn of the century and that the increase accelerated after 1909. The simultaneous increase and de- crease of two sequential month-intervals is a point of importance in relation to the period of greatest tree growth within the season. No. 18 TREE GROWTH AND RAINFALL—GLOCK 25 Table 12 gives the average year-to-year variation of rainfall ar- ranged in two parts, the first of which sets out the variations of March-July and January-August rainfall for the period 1909-1941 at the several stations. In view of the differences in elevation, the average variation of the rainfall is of the same order of magnitude TABLE 12.—Rainfall Average year-to-year variation March-July January-August 1909-1941 1909-1941 CHACON: 5 warts se ation vs ene 0.36 0.26 Blackpake ees uics a can wk sce 0.38 0.29 MEAS ercya tc Salte mile markets eno soe 0.36 0.25 miaosm Canyon @../2 dec bcancecuce 0.28 0.22 HB ASE VED AG Hen hee sie ete cha es eae tes 0.41 0.34 SoMtAR Neg teeta dione cet 0.35 0.27 PMID UIGIErque “ss 05 2h% os6 va eek 0.63 0.45 Santa Fe Albuquerque pees Bie pny 1850- 1898 1850- 1808- 1897 1941 1897 1941 IMarch=Jaly i... 66.0 cece e cess 0.42 0.33 0.79 0.60 FJamuary-Avicust 2.6... ccc 0.39 0.27 0.63 0.44 TABLE 13.—Rainfall Average departures March-July January-August 1909-1941 1909-1941 SACO gl ah ei lacs aaa whats 0.22 0.17 Blacisleakeserren et icc eer 0.25 0.19 MRO Suter siecle aerets Sioa Sesioaeloe exe 0.27 0.20 MAGS Canny Ol Px aeeeh aa 6 Soe cvartxs 0.20 0.17 LAG \WiteiGie pak Saas coon nes eoe 0.30 0.26 Sanitamleen tre errs eae 0.24 0.20 MNibuatienduemee sates sos eee eee 0.43 0.31 Santa Fe Albuquerque CSS EN (a 1850- 1898- 1850- 1898- 1897 1941 1897 1941 Mircticha=siialiya evar sci & «ior ouags'sxsieeers 0.33 0.22 0.58 0.40 anuaryoAnoust ...-..6-5..5.- 0.30 0.19 0.45 0.30 as that of the Holman Pass trees. The second part of the table shows the average variation of 1850-1897 to be greater than that of 1898- 1941, which is no doubt to be expected because of the lower average rainfall of the earlier period. Even the interval January-August at Santa Fe has the same decrease in the later period. Table 13 gives average rainfall departures. In the first portion of the table the departures for the rainfall of March-July and January- 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIT August during the period 1909-1941 are given for the several stations. These average departures are noticeably less than the comparable values for the growth layers of the Holman Pass collection. In the second portion of the table the average departures of the period 1850-1897 are distinctly higher than those of the period 1898-1941. The contrast between the two periods stands thus: a lower average TABLE 14.—Holman Pass tree growth and Santa Fe rainfall—characteristics 1850-1897 1898-1941 Average magmtude Rainfall) Giches)\i ainsi ase sae ona 6.34 6.74 Growth layers (mm.) Gags) a rein tence teal eR ear erence 1.53 1.70 1 RAP a OSM IRND trai aeeecrihctes ig Ue LAC 1.96 PAGrestiicted)) sesame cei acioe 1.99 1.83 Reainitall pc eee reer veel s hyo ero em ee 0.42 0.33 Growth layers GATOS ics wikiene ye aS gra aise omer 0.26 0.40 TID isso aerontcwet Merah eh ei 0.38 0.27 ANCTEStKICtEM)) naerseieeisiee eit 0.33 0.32 Rainiall eager cceee cece ee 0.33 0.22 Growth layers GeSlOn eee thence coma eae 0.32 0.36 Tike ecg Sete aunt ret eye meena 0.34 0.30 PICHESERICCEMI)) caloeatac cision aire oe 0.35 0.31 Raingallls ¥ sey zijeriecsyasusrenn orton 0.25 0.22 Growth layers GTO aes sarees sam cltee eee 0.132 0.148 DG es rest cise a Tome 0.204 0.118 Fis (GEStHICted ye mnete se 5,1 eae 0.130 0.105 rainfall during the earlier period is accompanied by a higher average variation and by a higher average departure. On the whole, such characteristics are to be expected.’° There remains, then, a comparison between the characteristics of the growth layers and those of rainfall for which the records of Santa Fe are used because of their length and continuity. Table 14 makes 10 Mixer, C. A., The rainfall year, Bull. Amer. Meteor. Soc., vol. 15, pp. 22-23, 1934; Williamson and Clark, Variability of annual rainfall in India, Geogr. Rev., vol. 21, pp. 675-676, 1931. Ss SS No. 18 TREE GROWTH AND RAINFALL—GLOCK 27 these comparisons using March-July rainfall. An increase in average rainfall (as between 1850-1897 and 1898-1941) is accompanied by an increase of average growth-layer thickness on dry sites (group 10) and by a decrease on wet sites (group 11); an increase in rainfall giving a decrease in its average variation is accompanied by an increase of average variation among growth layers from dry sites and by a decrease among growth layers from wet sites; an increase in rainfall giving a decrease in its average departure is accompanied by an in- crease of average departure among growth layers from dry sites and by a decrease among growth layers from wet sites; and an increase in rainfall giving a decrease in its average departure from mean variation is accompanied by an increase of average departure from mean varia- tion among growth layers from dry sites and by a decrease among growth layers from wet sites. In other words, changes of average variation, average departure, and average departure from mean variation among growth layers from wet sites follow the changes of the same features in the rain- fall whereas the growth layers from dry sites react in the opposite direction. The case is reversed as regards changes in growth-layer thicknesses: the growth layers from the dry sites follow the changes in average rainfall amounts. However, as discussed under the study of growth layers, two trees, HPC 1 and 7 of group Io, determine the characteristics of the group. Their elimination from the complete record leaves a homogeneous group consisting of the other two trees of group 10 and all trees of group 11 (HPC 8 is not included because its sequence is too short). This group of six trees, group 7 (re- stricted), is conformable within itself, and the changes in its charac- teristics from 1850-1897 to 1898-1941 agree with those of Santa Fe rainfall. Nevertheless, there remains the problem of why the average vari- ation of the dry-site trees increased with a decrease of average varia- tions in rainfall during 1898-1941. Calculation of the average varia- tion of rainfall for different months and month-intervals shows that only April and May increased their average variations during the period 1898-1941. It might be, therefore, that the dry-site trees were more influenced by April-May, or spring, rainfall than the wet-site trees were. Or the problem may concern limiting factors and optimum or near optimum soil-moisture conditions in that the wet-site trees grew under conditions where the trees responded directly and con- sistently to changes in rainfall. to io/2) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. III It is clear that the relations of tree growth to rainfall are highly complex not by themselves alone but also by the interplay of the entire range of growth factors, a circumstance emphasized by plant physiologists and ecologists.‘ The problem calls for much more work—it is far from finished. The observational method of field se- lection and laboratory analysis remains secondary to direct experiment on growing trees under controlled conditions. In any event, the present work comparing tree growth and rainfall characteristics strongly suggests that trees selected from certain sites and from the proper rainfall or soil-moisture zone can be sensitive indicators of rainfall changes through the use of average variation, average departure, and average departure from mean variation. CORRELATION OF TREE GROWTH AND RAINFALL Range of tests—Fairly extensive correlations were made between tree growth and Chacon and Santa Fe rainfall both for groups and for individual trees. The trend method was applied in its complete form until it was determined that variations of I or 2 years did not unduly distort the results. In addition to the more extensive corre- lations, selected tests were made between tree growth and the records of other rainfall stations. Tree growth (groups) and Chacon rainfall—Groups 1 to 9 were correlated with Chacon rainfall for the month-intervals shown in table 15. However, the table includes only those groups which were most significant. The table shows that correlations with March-July and January- August rainfall are the highest, and of these two intervals March-July is the more important. July rainfall is necessarily included as is indicated by the lower correlations of March-June. Of the 5 months, March-July, the rainfall of May-June is more important to tree growth than that of March-April and the rainfall of April is of less importance than that of May, June, or July. Apparently tree growth, as represented by the trees selected, responds directly to the rain which falls during and the several weeks immediately preceding the actual growing season. The most striking fact obvious at first sight is the correlation be- tween group 7 (all trees) and March-July rainfall, the trend coeffi- cient being 0.965 and the ratio of opposed trends 0.12. A trend ratio of 0.12 means that the trend of rainfall was opposite to the trend 11 Bot. Rev., vol. 7, pp. 651-655, 1941. No. 18 TREE GROWTH AND RAINFALL—GLOCK 29 of tree growth for 4 years out of 32 of variation. Of these opposite trends that for 1910 contains 84 percent of the numerical disagree- ment and, if 1910 be eliminated, the trend coefficient rises to 0.994. A comparison of the rainfall among all seven stations with the tree growth of group 7 for the 4 years of opposite trend, 1910, 1912, 1913, TABLE 15.—Correlation of tree groups and Chacon rainfall Trend coefficients and ratios of Opposed trends 1909-1941 G4 Gis G7 Go G 10 Git ioe! bt i 0.93 0.82 0.965 0.88 0.06 0.92 (0.22) (0.31) (0.12) (0.19) (0.16) (0.22) GO \ (a 0.92 0.73 0.05 0.88 0.04 0.91 (0.16) (0.31) (0.19) (0.25) (0.22) (0.28) Mar Jie <....... 0.85 0.82 0.89 0.77 (0.28) (0.28) (0.16) (0.28) MaysAug. .s....6.% 0.80 0.68 0.80 0.66 (0.25) (0.37) (0.28) (0.28) May-June ......... 0.73 0.70 0.72 0.49 (0.31) (0.34) (0.28) (0.34) Wam—May os... acs 0.72 0.32 0.70 0.62 (0.25) (0.44) (0.28) (0.41) Neyo May