: eae 2 2 “ EPS z SSS aie _ am -~ _ se s ~ - -- = oa S cS SMITHSONIAN MISCELLANEOUS COLLECTIONS VOR! Seeeaecee® “EVERY MAN IS A VALUABLE MEMBER OF SOCIETY WHO, BY HIS OBSERVATIONS, RESEARCHES, AND EXPERIMENTS, PROCURES KNOWLEDGE FOR MEN’’—JAMES SMITHSON (PusricaTIon 4310) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION 1958 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U.S. A. 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 An- nual 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. LEONARD CARMICHAEL, Secretary, Smithsonian Institution. (iii) IO. ET. CONTENTS . Aszot, C. G. Leading operations of the Smithsonian Astro- physical Observatory, 1895 to 1955. 8 pp. Sept. 22, 1955. (Publ. 4222.) Peterson, Menpet L. The last cruise of H.M.S. “Loo.” 55 pp., 17 pls., 3. figs. Nov. 23, 1955. (Publ. 4224.) FaircHILp, G. B. Synonymical notes on neotropical flies of the family Tabanidae (Diptera). 38 pp. Jan. 11, 1956. (Publ. 4225.) Coorer, G. ArtHur. New Cretaceous Brachiopoda from Ari- zona. 18 pp., 4 pls. Dec. 21, 1955. (Publ. 4227.) WetMorE, ALEXANDER. A check-list of the fossil and prehis- toric birds of North America and the West Indies. 105 pp. Jan. 25, 1956. (Publ. 4228.) Gazin, C. Lewis. Paleocene mammalian faunas of the Bison Basin in south-central Wyoming. 57 pp., 16 pls., 2 figs. Feb. 28, 1956. (Publ. 4229.) Gazin, C. Lewis. The upper Paleocene Mammalia from the Almy formation in western Wyoming. 18 pp., 2 pls. July 31, 1956. (Publ. 4252.) Gazin, C. Lewis. The geology and vertebrate paleontology of upper Eocene strata in the northeastern part of the Wind River Basin, Wyoming. Part 2. The mammalian fauna of the Bad- water area. 35 pp., 3 pls., 1 fig. Oct. 30, 1956. (Publ. 4257.) Kiruam, Lawrence. Breeding and other habits of casqued hornbills (Bycanistes subcylindricus). 45 pp., 6 pls., 2 figs. Nov. 8, 1956. (Publ. 4259.) Snoperass, R. E. Crustacean metamorphosis. 78 pp., 28 figs. Oct. 17, 1956. (Publ. 4260.) Cuapwick, L. E. The ventral intersegmental thoracic muscles of cockroaches. 30 pp., 18 figs. Jan. 15, 1957. (Publ. 4261.) (v) aret¢ Oe) ee ME Se! ; oc aed) fF WM nd sd a 1 pee sh ne oe ee ee ss =~ fm: aT att) Mh a ee - 41a This ef, ds) Gena ete wa) Sar) sie =o SEN PSs & | Oe ogy) tal) Se ~ oat ei 78 aaa ae eee he = 7 r mets miner «2 44 4, shay filth een’ ied | : (ey Pts Sah’ > Vie a ae : 1 lan - : *e (Dei Ge of ' ee ye e 7 w 7 ae 7 7 7 - a a “ . Ree. aa ar a ¢q e ve) ee I : > a - Jen & i 6. Gite SA85 2 Ei a 2 es - - i tS tah fall hope igpmwait lie ii ean ee gt) I bas a | sper | nares ae ra or. - ha oy i SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 1 Roebling Fund LEADING OPERATIONS OF THE SMITHSONIAN ASTROPHYSICAL OBSERVATORY, 1895 to 1955 By Cc. G. ABBOT Research Associate, Smithsonian Institution CO! Saar Pe iS OC Wows 2 (PUBLICATION 4222) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION SEPTEMBER 22, 1955 The Lord Baltimore Press BALTIMORE, MD., U. 8. A, Roebling Fund LEADING OPERATIONS OF THE SMITHSONIAN ASTROPHYSICAL OBSERVATORY, 1895 TO 1955 By C. G. Aszor Research Associate, Smithsonian Institution INTRODUCTION Having been associated with the Astrophysical Observatory almost from its inception in 1890, it seems good to me to print, in small com- pass in one place, references to the leading researches and instrumental developments carried out there. It may well be that there are now, and will be in future, those who, for one reason or another, may wish to refer to these events, and will appreciate having easy access to the original sources. The list is far from exhaustive, either as regards the work of the Astrophysical Observatory, or references to it. But I believe it is suffi- cient to present a fair picture of what has been accomplished. SECTION A Part 1.—lmproved and new instruments 1. The bolometer rebuilt and equipped Annals of the Astrophysical Observa- with a balancing device close be- tory,) vol. I, pp. 47-56, 105-109, side it and at constant tempera- 1900; vol. 3, p. 42, 1913. ture. Result: The drift nearly eliminated and the wiggle greatly reduced. Galvanometer. Theory investi- Astrophys. Journ., vol. 18, No. 1, July gated and new galvanometer of 1903. tenfold sensitiveness built. Annals, vol. 1, pp. 244-252, 1900. 3. Vacuum bolometer with self-con- Annals, vol. 4, pp. 45-64, 1922. tained Wheatstone bridge built. Result: Several-fold increase of sensitiveness, and increased i steadiness. 4. Silver-disk pyrheliometer invented. Smithsonian Misc. Coll., vol. 56, No. About 100 copies have been 19, IQII. 1 Hereafter referred to simply as “Annals.” SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL 131, NO. 1 10. It. 12, 13. 14. 15. 16, SMITHSONIAN MISCELLANEOUS COLLECTIONS built, standardized, and sold at cost to observers throughout the world. Repaired at cost and restandardized free when damaged. Water-flow and water-stir stand- ard pyrheliometers invented and used. The world’s scale of solar radiation measurements rests on them.? Pyranometer invented. Used daily on short-method solar-constant observations. Used by Moore in North Carolina and Chile. Copies made and sold at cost world-wide. Honeycomb pyranometer, or meli- keron, invented. Used by Ab- bot and Aldrich on human body, and by Sverdrup in polar re- gions. Copies made and sold at cost world-wide. Balloon recording pyrheliometer invented and used at high alti- tudes. Two-mirror coelostat invented. Slide-rule extrapolator invented. Constantly used in long-method solar observing. High-power lamp and other de- vices prepared by F. E. Fowle for researches on deep infrared spectrum. Highly sensitive radiometer in- vented for measuring energy spectra of stars. A prism of nearly normal wave- length dispersion invented. The kampometer invented, a highly sensitive instrument for measuring radiation. The periodometer invented, a me- chanical instrument for discov- ering periodic changes in data. A multiple rotating-sector dia- VOL. 131 Annals, vol. 3, pp. 47-52, 1913; vol. 7, p. 105, 1954. Smithsonian Misc. Coll., vol. 95, No. 23, 1937; vol. 111, No. 14, 1949. Annals, vol. 3, pp. 52-72, 1913; vol. 7, Pp. 99-101, 1954. Smithsonian Misc. Coll., vol. 87, No. 15, 1932; vol. 110, No. 11, 1948. Smithsonian Misc. Coll., vol. 66, No. 7, 1916. Annals, vol. 4, pp. 65-84, 1922; vol. 7, pp. 15-16, 21, 138, 1954. Smithsonian Misc. Coll., vol. 72, No. 13, 1922. Annals, vol. 4, pp. 41, 300, 1922; vol. 5, PP. 43-45, 1932. Smithsonian Misc. Coll., vol. 65, No. 4, 1915. Annals, vol. 4, pp. 347-365, 1922. Annals, vol. 2, pp. 22-23, 211, 1908. Annals, vol. 4, pp. 84-86, 1922. Annals, vol. 4, pp. 23-25, 274-287, 1922. Astrophys. Journ., vol. 69, pp. 293-311, 1920. Smithsonian Misc. Coll., vol. 104, No. 14, 1945. Astrophys. Journ., vol. 11, No. 2, pp. 135-139, March 1900, Smithsonian Misc. Coll., vol. 104, No. 22, 1946; vol. 107, No. 19, 1948. Smithsonian Misc. Coll., vol. 89, No. 3, 1933. Smithsonian Misc. Coll., vol. 87, No. 4, 1932. Annals, vol. 5, p. 96, 1932. eee * A.P.O. modified form of Angstrém pyrheliometer is used in daily observa- tions. See Annals, vol. 6, pp. 50-55, 1942. NO. I ASTROPHYSICAL OBSERVATORY, 1895 TO I955—-ABBOT phragm combination invented, instantly exchangeable, for bo- lometer work. 17. A continuously variable rotating sector invented, of accurate ratio, for photometry. 18. A pair of telephoto cameras in- vented, electrically connected, for simultaneous exposure on flying objects. The invention comprises a_ belt-focal-plane shutter, surrounding film spools. Shutter and spools operated by a long spring and clockwork. The observer and assistant sepa- rated by a measured base line keep both cameras trained. Ob- server makes a series of expo- sures by a trigger, and second camera is simultaneously exposed. 19. Apparatus invented for prevent- ing “personal equation” in ob- serving sudden phenomena. The observer notes the sector where, not the times when, the event occurs. 20. Automatic recording radiation instruments invented. No published description. One camera on public exhibition in Langley case in the West Hall of the Arts and Industries Building, Smithsonian Institution. Apparatus Langley case (see above). Annals, vol. 7, pp. 144-146, 1954. on public exhibition in Part 2.—Various inventions, mainly for military use in World Wars I and II I. 2. Variable-speed governor. Variable-speed power-transmis- sion mechanism, Claim 1, al- lowed “The combination of a driving element, a driven ele- ment, and means for establish- ing, and maintaining constantly, exactly and positively, a desired speed ratio between said ele- ments, or for continuously vary- ing said ratio.” For a clockwork to be of speed varied at will, without stopping, and continuously, through a several- fold range. Used for a Navy project. 3. Self-propelled rotating projectile for smooth-bore guns, Combina- tion with smooth-bore ordnance. U. S. Patent No. 893416 of July 14, 1908. U. S. Patent No. 2367254 of Janu- ary 16, 1945. U. S. Patent No. 1380172, and U. S. Patent No. 1380171, both of May 31, 192I. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I 4. Gyroscopic navigation instrument. R 6. For measuring differences in longitude and latitude without sun or star observations. Compass and magnetic-dip indi- cator. Both this and No. 4 used the principle of neutral flotation in liquid, and electric current therethrough for operat- ing. Germans independently dis- covered the mathematical prin- ciple of No. 4 and built such a machine but it failed. An Eng- lishman from National Labora- tory examined patent of No. 4, and said it carried superior fea- tures. Work on it stopped with the Armistice, November 1918. Instrument for navigating air- planes by daylight star observa- tions. Stars can be seen with a small telescope in daylight if the telescope field contains the star image. The instrument could be set to contain the star in its field before observing. Twelve stars and two planets were easily observed by W. H. Hoover in New Mexico. E. D. McAlister observed Altair from airplane at 21,000 feet. Instrument for automatic mapping of airplane course over ocean, to enable return to course of mother ship. The patent, No. 2367254, above cited, was a part of this device. Solar distilling apparatus. Solar heater. Solar heat collector. General Electric Patent No. 1501886 to C. G. Abbot July 15, 1924. General Electric Patent No. 1533683 to C. G. Abbot April 14, 1925. Built and tested secretly. Never pub- lished. Built and tested secretly. Never pub- lished. Patent No. 2141330, December 27, 1938. Patent No. 2247830, July 1, 1941. Patent No. 2460482, February 1, 1949. SECTION B Part 1.—Researches . Bolometric map of infrared solar spectrum. Dispersion of rock-salt and fluo- rite. (Six-place decimals in re- Annals, vol. 1, pp. 5-204, 1900; vol. 5, P- 54, 1932. Smithsonian Misc. Coll., vol. 82, No. 1, 1929. Annals, vol. 1, pp. 219-237, 253-262, 1900. NO. 10. Il. 12, 1g: 14. 15. 16. 17. 18. 10. I ASTROPHYSICAL OBSERVATORY, 1895 TO I1955—-ABBOT 5 fractive index called ridiculous by Holland physicists. Identical in fifth place with Paschen work, however.) Structure of water-vapor bands Wy and Ws. Total solar eclipses, 1900, I9QoI, 1908, 1918, 1910. Theory of sensitive galvanometer. “The cheapest form of light.” Solar-constant and solar-distribu- tion work, begun in 1902. Mount Wilson expeditions, begun 1905. Theory of atmospheric transmis- sion. Methods for measuring the solar constant. Transmission of the spectrobo- lometer. Pyrheliometry. Details of solar-constant observ- ing. Sources of error in solar-constant work. Solar-contant results of stations compared. Normal solar-energy curves. Pre- ferred determination. Sun’s temperature. Atmospheric transmission, many stations, sea level up to 14,000 feet altitude. Theory of vacuum bolometer, cor- rected later. Annals, vol. 1, pp. 263-264, 1900. Astrophysical Observatory special eclipse volume, 1900. Annals, vol. 2, p. 2, 1908; vol. 3, pp. 3-6, 1913; vol. 4, PP. 20, 31, 34, 35, 1922. Smithsonian Misc. Coll., vol. 69, No. 9, IQI9. Annals, vol. 1, pp. 244-252, 1900. Astrophys. Journ., vol. 18, No. 1, July 1903. Annals, vol. 2, p. 5, 1908. Annals, vol. 2, pp. 2, 3, 21-82, 211-228, 1908. Annals, vol. 2, pp. 7, 83-116, 1908. Annals, vol. 2, pp. 13-17, 1908. Annals, vol. 2, pp. 17, 57, 117-124, 1908. Annals, vol. 2, pp. 24, 51, 52, 1908. Annals, vol. 2, pp. 34-49, 1908; vol. 3, PP. 47-72, 1913; vol. 7, pp. 21-23, 1954. Annals, vol. 3, pp. 21-20, 1913; vol. 6, pp. 43-81, 1942. Annals, vol. 2, pp. 58-82, 1908; vol. 4, pp. 161-176, 1922; vol. 5, pp. 110-131, 1932; vol. 6, pp. 33-42, 1942. Annals, vol. 2, pp. 85-98, 1908; vol. 3, p. 134, 1913; vol. 4, pp. 177-182, 1922; vol. 5, pp. 244-245, 1932; vol. 6, p. 163, 1942. Annals, vol. 2, pp. 104-106, 1908. Smithsonian Misc. Coll., vol. 74, No. 7, 1923. Annals, vol. 2, pp. 106-107, 1908; vol. 3, PP. 194-201, 1913. Annals, vol. 2, pp. 96-98, 110-112, 1908; vol. 3, pp. 104-113, 1913; vol. 4, Pp. 131-158, 1922; vol. 5, pp. 168- 193, 1932; vol. 7, pp. 95-98, 1954. Annals, vol. 4, pp. 45-64, 1922; vol. 5, pp. 75-81, 1932. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS 20. Infrared and ultraviolet correc- tions for solar-constant work. 21. Solar variation: a. First suspected. b. Clayton’s contributions. c. From solar-constant work 1920-1930. d. Short up and down trends and (1) temperatures, (2) ionosphere. e. Accompanying (1) hurri- canes, (2) magnetic storms, f. Accompanying sunspots. g. Periodic—(1) 27-day, (2) 6.6485-day. h. Long periodic and weather. 22. A large family of periodic varia- tions: a. In the sun. b. In the weather. 23. Defense of our solar-constant value (Abbot, Fowle, Aldrich). 24. Brightness of the night sky. 25. Direct and scattered radiation of sun and stars. 26. Tower telescope on Mount Wilson and solar-drift curves. 27. Nature of the sun’s sharp boundary. 28. Volcanoes and climate. 29. Summary of the work of the As- trophysical Observatory, 1890- 1920. 30. Radiometer measurements of stel- lar-energy spectra. VOL. 131 Annals, vol. 5, pp. 103-110, 1932. Astrophys. Journ., vol. 19, p. 305, June 1904. Annals, vol. 2, pp. 98-103, 117-179, 1908. Annals, vol. 4, pp. 36, 185, 367-374, 1922. Annals, vol. 5, pp. 246-269, 1932. (1) Smithsonian Misc. Coll., vol. 95, Nos. 12 and 15, 1936; (2) vol. 104, No. 13, 1945. (1) Smithsonian Misc. Coll., vol. 110, No. 1, and (2) No. 6, 1948. Smithsonian Misc. Coll., vol. 110, No. 6, 1948. Annals, vol. 7, pp. 165-168, 1954. (1) Smithsonian Misc. Coll., vol. 104, No. 3, 1944; vol. 116, No. 4, 1951; (2) vol. 111, No. 13, 1949. Smithsonian Misc. Coll., vol. 122, No. 4, 1953. Smithsonian Misc. Coll., vol. 122, No. 4, 1953. Smithsonian Misc. Coll., vol. 128, No. 4, 1955. Smithsonian Misc. Coll., vol. 128, No. 3, 1955. Annals, vol. 4, pp. 323-366, 1922. Astron. Journ., vol. 27, No. 3, pp. 17- 24, June 20, IQII. Astron. Journ. vol. 28, No. 16, pp. 129- 135, March 1914. Annals, vol. 4, pp. 217-257, 1922. Smithsonian Misc. Coll., vol. 78, No. 5, 1926. Scientia, vol. 19, pp. 171-181, March 116. (See also Abbot, C. G., “The Sun,” IQOII.) Smithsonian Misc. Coll., vol. 60, No. 29, 1913; vol. 65, No. 9, 1916. Annals, vol. 5, pp. 1-5, 1932. Astrophys. Journ., vol. 50, pp. 87-107, 1924. Astrophys. Journ., vol. 69, pp. 203-311, 1920. NO. I ASTROPHYSICAL OBSERVATORY, 1895 TO 1955—-ABBOT 7 31. Campaign of observations of solar intensity on surfaces of different orientations and with various spectral regions, made at army camps for Quartermaster Corps, for a period of 8 years. 32. Daily solar-constant values, 1920- 1952, with 10-day and monthly means. 33. Convenient table for solar-constant tabulations. 10-day and monthly mean excesses over 1.900 in hundredths percentages of 1.94. Thus 1.950 becomes 1.950-1.900 1.04 s 100 = 2:58: Similarly 1.940 becomes 2.06. This difference, 0.52, is 0.53 per- cent of mean solar constant. Nore.—The tables in the two references cited above are printed without the decimal point for economy, and do not correspond with the descrip- tions above unless this fact is known. Annals, vol. 7, pp. 144-164, 1054. Annals, vol. 5, pp. 177-182, 1932; vol. 6, pp. 85-162, 169-175, 1942; vol. 7, pp. 26-94, 1054. Smithsonian Misc. Coll., vol. 117, No. 10, pp. 20-24, 1952; vol. 128, No. 4 (table 1), 1955. Part 2.—Work of spectalists te 1b Be Aldrich: a. The melikeron, an approxi- mately black-body pyra- nometer, b. Reflecting power of clouds, and earth’s albedo. c. Eclipse expedition, June 1918. d. A study of body radiation. e. Sunspots and the solar con- stant. f. Various researches on long-wave rays. g. Author (with W. H. Hoover) of volume 7 of Annals of the Astro- physical Observatory. 2. F. E. Fowle: a. On atmospheric precipita- ble water. Smithsonian Misc. Coll., vol. 72, No. 13, 1922. Annals, vol. 4, pp. 375-381, 1922. Smithsonian Misc. Coll., vol. 69, No. 9, 1919. Smithsonian Misc. Coll., vol. 81, No. 6, 1928. Annals, vol. 7, pp. 165-168, 1954. Annals, vol. 4, pp. 287-299, 1922. Annals, vol. 7, 1954. Astrophys. Journ., vol. 35, p. 149, 1912, 3: SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 b. On Avogadro’s number. c. On atmospheric ozone. d. On water-vapor absorption above 3 microns. e. On water-vapor absorption below 3 microns. f. Preparation of Physical Tables. W. H. Hoover: Besides his large part in vol- ume 7 of the Annals of the Astrophyical Observa- tory, as coauthor with L. B. Aldrich, he engaged in clas- sic researches on photosyn- thesis as a member of the staff of the Division of Radiation and Organisms, later a branch of the Astro- physical Observatory. a. Carbon-dioxide assimilation in a higher plant (with Earl S. Johnston and F. S. Brackett). b. The dependence of carbon- dioxide assimilation in a higher plant on wave- length of radiation. c. Improvements in use of standard water-flow pyr- heliometer, and in silver- disk pyrheliometer. d. Special studies of global sun and sky radiation (with L. B. Aldrich). e. Mechanical integrator for Brown recording poten- tiometer. Astrophys. Journ., vol. 40, p. 435, 1914. Smithsonian Misc. Coll., vol. 81, No. II, 1920. Annals, vol. 3, pp. 171-193, 1913. Annals, vol. 4, pp. 274-287, 1922. Smithsonian Physical Tables, 5th ed., 1910; 6th ed., 1914; 7th ed., 1919; 8th ed., 1934. Smithsonian Misc. Coll., vol. 87, No. 16, pp. I-19, January 16, 1933. Smithsonian Misc. Coll., vol. 95, No. 21, pp. I-13, February 27, 1937. Smithsonian Misc. Coll., vol. 122, No. 5, pp. 1-10, August 14, 1953. Annals, vol. 7, pp. 99-104, 1954. Annals, vol. 7, pp. 144-164, 1954. Annals, vol. 7, pp. 138-139, 1954. ot re : eT i ai Laer ie ; s T ul ni rw i na i) Le 7 a ne “ 1 NOD 2, PEL OLLECTIONS wn MISCELLANEOU SONIAN MITH [PABN OY} Worl Y “TQTPBUIS ATUSTS Ys o 10Ud ) (‘purl suy “USIMUDATL) “UINOSN YAY ‘0O'T JY} SB SSP]D OURS dYy JO diys e@ Jo soainjord 30 sjapouwl suljsrxa Ou 91k oTOYT ouqye adAy SIY} OF Aeypruts K1OA SBM 0oT MUL ‘OIZI ynoqe yo diys uns-o§ Ysiiig eB Jo Japoy, REECE — ak - ———S—S——— } SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 2 Mee As? CRUISE OF Ti Ms, “LOO” (WirtH 17 PLatEs) By MENDEL L. PETERSON Curator of Naval History U. S. National Museum Smithsonian Institution i H5ONS* 1 a Sea S220000000% (PuBLIcATION 4224) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION NOVEMBER 23, 1955 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U.S. A. bhahebASY CRUISE: OF ELMS. shOO" By MENDEL L. PETERSON Curator of Naval History U. S. National Museum Smithsonian Institution (Wirt 17 PLateEs) In the spring of 1951 I was invited by Dr. and Mrs. George Crile, Jr., of Cleveland, Ohio, to accompany them on an expedition to ex- plore remains of a ship that had been wrecked about 5 miles off the main line of the Florida Keys over two centuries ago. The wreck had been shown to them by William Thompson, of Marathon, Fla. Only one thing was known about the ship—that it went down sometime after the year 1720. This was indicated by the fact that in 1950 Dr. and Mrs. Crile and their party had recovered some copper coins from the site, one of which was a Swedish half-ore piece (pl. 2, fig. 1) dated 1720. The site was a reef named “Looe” on the charts and was located some 25 miles southwest of Marathon. At the time, the source of the name was unknown, and its presence on the charts was not considered significant. On Sunday evening, May 27, most of the members of the expedi- tion assembled in Miami, and the next day left for the Keys and Thompson’s yacht harbor, which was to be the base of operations. Here the entire party came together. It consisted of the sponsors, Dr. and Mrs. Crile; Mr. and Mrs. E. A. Link, of Binghamton, N. Y.,? Mr. and Mrs. James Rand, of Cleveland, Ohio; Mr. and Mrs. John Shaheen, of New York City ; William Thompson, of Marathon, Fla. ; Arthur McKee, of Homestead, Fla., an experienced diver on ship- wreck sites; and myself. Necessary supplies and equipment were as- sembled and tested, and the boats were made ready. We were to use a barge built on a Higgins boat hull, a small fishing launch, and later Mr. Link’s yawl, the Blue Heron. Early Wednesday morning, May 30, the party left for the reefs 1 The participation of Mr. and Mrs. Link in the expedition was to prove very fortunate for the National Museum since it led to the establishment of the Link Fund through their generosity. This fund enables the Museum to participate in annual expeditions to explore historic wreck sites in the Florida Straits area. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 131, NO. 2 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I and by midmorning were over the wreck site (pl. 3). To locate the wreck exactly it was necessary to get into the water with face plates and carefully scan the sand bottom of a “valley” lying between two fingers of the reef which ran out to sea. Only the metal objects from the ship remained, and these were covered with a sand crust giving them the same color as the bottom—excellent camouflage, which made them almost invisible from the surface of the water. We detected the outlines of bars of metal, the ring of one of the ship’s anchors (pl. 4) jutting from the reef, two long cylindrical objects, encased in marine growths, which were recognized as guns from the ship, and, upon closer inspection, piles of round objects encrusted with sand—solid shot for the ship’s guns. After a preliminary survey from the surface, the boats were pulled over the wreck and the diving gear was prepared for operation. Tight-fitting rubber masks that completely covered the face were connected to the air compressors by long lengths of strong rubber hose. The divers strapped on their lead belts, put on their masks or helmets, and went over the side on the diving ladder and lifeline. The first object recovered was a large chunk of metal roughly tri- angular in cross section and stepped on the surface. It proved to be solid cast iron and was identified as permanent iron ballast cast to fit along the keelson of the ship. Clinging to it was a solid iron shot 33 inches in diameter, a standard 6-pound shot of the eighteenth century. These finds immediately revealed two additional facts concerning the ship—she was most probably a warship, since merchantmen carried disposable ballast of stone, and she had 6-pounder guns in her batteries. The next day the attention of the divers was devoted to the smaller objects lying about in the sand “‘potholes” on the site. By the use of a powerful jet of water the sand was carefully washed away and the articles were uncovered (pl. 5). Soon basketfuls of sand-encrusted hull bolts, nails, solid iron shot, fragments of rum or brandy bottles, Chinese porcelain dishes, and earthenware, and many other objects were being emptied on the decks of the salvage boats. On the first of two brief dives that the author made on the site a basketful of solid iron shot was gathered (pl. 6). In this lot was found a 6-pound shot with an arrow on it, which was immediately identified as the broad arrow (pl. 7). This was the first indication of the nationality of the ship, as this symbol has been used for centuries by the kings of England and Great Britain to mark royal property. The occurrence of the broad arrow on the shot was not conclusive evidence that the ship had been British, since ordnance stores could have been captured NO. 2 LAST CRUISE OF H.M.S. “LOO”—PETERSON 3 or stolen by the enemies of Britain. But until further evidence proved the contrary, we could consider the ship to have been British.? In the basket 12-pound, I-pound, and 4-pound shot were also found, giving additional information on the ship’s batteries. The broad arrow also appeared on the 12-pound shot. On Thursday, May 31, and Friday, June 1, numerous small objects were brought up, including more porcelain fragments, parts of clay pipes and rum bottles, the wooden knob of a walking stick, the eye- piece of a navigation instrument, pieces of stoneware decorated with blue flowers, and animal bones (later identified as pig and cow) from the pickled-meat stores of the ship. (See pls. 8, 9, and 10.) On Saturday, June 2, the party remained ashore to sort, clean, and begin the preservation process on the objects recovered. Fragments of wood were packed in fresh water for shipment to the National Mu- seum, the sand crust was cleaned from the cast-iron and other large iron objects by light hammering, and the objects were placed in baths of fresh water to leach out the sea salts. The cast iron was found to have been oxidized to a depth of one- half to three-fourths of an inch. The removal of the sand crust from all surfaces (those portions having been converted to crystalline mag- netite, which was very friable) had to be done with great care. The porous oxidized layer was saturated with chlorides, and to break these down the cleaned shot were placed in baths of sodium hydroxide. Most of the shot were packed with the sand crust on them, the crust protecting them from excessive drying while on the way to the Museum. On Sunday and Monday (June 3 and 4) a continual stream of ma- terial came up from the wreck and was added to the piles ashore at our base (pls. 11, 12). There was such a quantity of specimens that _it was decided that the author should devote a full day to identifying, sorting, cleaning, and preserving those that were to be retained. The boats went out as usual and that evening returned with one of the cannon barrels. Mr. Link had rigged the main boom of the Blue Heron and had lifted it to her side with block and tackle (pls. 13, 14). Through skillful seamanship and favorable weather the 2,000- pound barrel was brought to Marathon, hanging beside the delicate mahogany hull of the yawl, which was protected with rope fenders. As soon as the barrel was on the ground at our base we began re- moving the sand crust with a hammer. As the crust fell away (pl. 15) 2 Later a chain plate, which was originally bolted to the ship’s hull, was found by a salvage party from Miami. It also bore the broad arrow. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I the beautifully molded contours of an eighteenth-century barrel were revealed (pl. 16). The appearance was deceptive, however, for while the physical form of the barrel was perfect it was evident that the surface of the iron had been oxidized deeply. Two more clues to the identity of the ship were furnished by the barrel. On the second reinforce over the trunnions was a crowned rose, and in the muzzle were the remains of a wooden tompion. The crowned rose was an insigne employed by the Tudor and Stuart monarchs of England and was not used as the principal mark on royal artillery after the death of Queen Anne in 1714. The fact that the tompion was in the barrel indicated that the ship had run aground through accident or storm and not as an aftermath to a naval engage- ment. The crowned rose enabled us to estimate the date before which the ship sank. Since the normal life of an iron barrel on shipboard was usually not over 35 or 40 years, the barrel had probably not been in active service after 1714 plus 35 or 40 years, or 1749-1754. It was therefore assumed that the ship must have sunk before the year 1750. The barrel was the last important object recovered from the wreck site in 1951, and we now had all the evidence that we could expect to recover that year. This evidence had told us that the ship was a British warship, had sunk between 1720 and 1750, had 6- and 12- pounders in her main battery, and had gone down as the result of an accident and not as a sequel to a naval action. We knew, of course, that the reef on which the ship had sunk was called “Looe Reef,” but we had not suspected a connection between the name and the wreck itself. After my return to Washington I began a search of the ship casualty lists for the eighteenth century published in Clowes “The Royal Navy,” and found the entry—“‘1743 Looe 44 guns, Capt. Ashby Utting, Lost in America.”’ Further research indicated that she carried 6- and 12-pounders. The conclusion was obvious—the ship we had investigated was the Loo whose wreck had given her name to the reef. That night I phoned Dr. and Mrs. Crile in Cleveland and they immedi- ately called a friend in London. Within a week the Public Record Office had yielded a letter written by Captain Utting at Port Royal, S. C., February 15, 1743/44,° in which he described the wreck of his ship. A year later I was in London digging out all the documents in the Public Record Office relating to the ship. The account that fol- lows is based on letters, the Navy List, the Loo’s pay lists and muster 8 The legal year began March 15. The calendar year was 1744. (See Appen- dix B for Utting’s letter.) NO. 2 LAST CRUISE OF H.M.S. ‘‘LOO’”—PETERSON 5 rolls and other documents in the Public Record Office in London. All these documents are Admiralty papers. The references are given in the manner in which they are numbered in the collections of the Pub- lic Record Office, and bear the prefix ADM or AD. On June 14, 1743, Thomas Corbett, Secretary of the Admiralty, sat down in his London office and countersigned an order directing Capt. Ashby Utting to prepare his ship, the frigate Loo,’ for a cruise to North America: Having order’d His Majesty’s ship under your command to be refitted at Portsmouth, for a voyage to North America, cleaned, sheathed and graved,> and her provisions compleated to six months of all species, except beer, and of that as much as she can conveniently stow, and stored accordingly; you are hereby required and directed, to repair with her into Portsmouth Harbour, and strictly to observe the following instructions. You are to give constant attendance. Wie ie? Gis By TG: Thus began the last cruise of the Loo, the story of which might have been taken from a classic work of fiction. The Loo, a frigate of 40 to 44 guns, had seen long service in the Royal Navy. She had been built during the expansion of the British fleet incident to the War of the Spanish Succession. In this war England was fighting to prevent the seating of a Bourbon’ on the throne of Spain—a scheme of Louis XIV to strengthen the position of France in Europe. The Loo was to meet her end during another war in which Spain and Great Britain were enemies, a war that began as a result of the succession of Maria Teresa to the throne of Austria. 4 Named for the old seaport town of Looe (also Loo), which lies on the rocky coast of Cornwall and which has supplied sturdy sailors to the Royal Navy since its beginning. 5 The first Royal Navy vessel to be sheathed with lead was the Phoenix; this was done in 1670. The practice had been followed in the Spanish Navy since the middle of the sixteenth century and in some cases by English merchant ships (see Clowes, The Royal Navy, vol. 2, p. 240). Lead proved impractical, how- ever, and the practice of sheathing with thin fir boards was followed until the time of the American Revolution, when the British fleet was sheathed with copper. The thin fir sheathing was backed with pitch and horsehair, which dis- couraged worms from tunneling into the ship’s planking. Graving was the proc- ess of burning sea life from the bottom of a ship. ® Tnitials of the Lords of the Admiralty, “W” for Daniel, Earl of Winchelsea, First Lord of the Admiralty, March 19, 1742, to December, 1744. “T.C.” for Thomas Corbett, Secretary of the Admiralty, 1742-1751. (Admiralty Out- Letters, ADM 2, vol. 60, p. 15, Public Record Office.) 7 The grandson of Louis XIV, who ruled as Philip V of Spain (1700-1746). 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 She was most probably launched in 1706 since she first appears in the Navy List under the date April 1, 1707, when she was at “Long- reach taking in Guns.” ® Her complement at that time is given as 190 men and her battery as 42 guns. The Navy List thereafter follows her career in terse monthly entries (see Appendix A). Her first two cruises carried her to Archangel and Newfoundland.?° By 1709 she was back at Sheerness refitting, and afterward was on patrol duty in the Channel.‘ Early in 1710 she was attached to the Dunkirke Squadron and in the fall of that year was on convoy duty to the eastern countries.’* In April 1711 she was in Holland “to bring the Queen’s wine to the Nore.’”’** She then sailed convoy to Russia and during the last two months of 1711 was refitted and again sent to duty in British waters, meeting ships from Virginia and convoying them to British ports.4* During the winter of 1712-13 she transported troops to Bayonne and returned with prisoners of war and then again served in the Channel, cruising against smugglers.1* In the fall of 1714 she was sent convoy to Port Mahon in the Mediterranean and on return was paid off and laid up."7 Three years later the Loo was back in service as a hospital ship on duty with the Baltic Squadron.’* She was then laid up for the winter and the next spring again commissioned as a hospital ship and sent to the Mediterranean Squadron. A year later (April 1719) she appears in the Navy List with 30 guns and a crew of 125 men, which probably indicates that she had been reconverted to a frigate.1° Thus fitted out she served with the Mediterranean Squadron until the spring of 1722.°° From that time until January 1728 she appears to have been laid up. On January 10, 1728, she was again in sea pay and until 8 At least one earlier Loo is recorded. Like her successor, she was a frigate of 40 guns and was also lost through shipwreck, having run aground on the Irish coast near Baltimore, April 30, 1697 (Clowes, The Royal Navy, vol. 2, p. 536). ® Navy List, Jan. 1, 1707-Dec. 31, 17090, ADM 8/10, Public Record Office. 10 ADM 8/10. 11 Thid. 12 Thid. 13 Thid. 14 ADM 8/10 and 8/11. 15 ADM 8/11. 16 ADM 8/12. 17 bid. 18 ADM 8/13. 19 Tbid. 20 Tbid. NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 7 July 1730 was on duty in British waters performing such tasks as transporting clerks and money to the pay at Plymouth and patrolling the Channel.?* In August she was ordered to the Mediterranean as convoy for transports going to Gibraltar and remained in the Medi- terranean cruising against the Barbary pirates “on the Coast of Sallee’ Coming home to Britain in August 1731, the Loo was again on Channel service until the next spring. For three years she was again laid up and on May 5, 1735, was commissioned and fitted out as a hospital ship for duty in the Channel service. In August of that year she joined the naval forces at Lisbon, still as a hospital ship, and served there until the spring of 1737.°* From that time until January 1742 she was laid up and, war having broken out between Great Britain and Spain, was recommissioned as a frigate of 44 guns on January 5 and placed in the Channel service under the command of the Earl of Northesk. While on a cruise in the area of Cape Finnisterre (northwest Spain) the Loo, in company with the Dealcastle (24 guns) raided Vigo Bay, capturing four Spanish vessels in the harbor, an incident reported in the London Gazette for August 31, 1742.7 21 Tbid. 22 Thid. 23 ADM 8/17 and 8/18. 24 ADM 8/t19 and 8/20, 25 Also mentioned in the Gentlemen’s Magazine for August 1742, p. 445, and September 1742, p. 494, giving an account of the Loo raiding in the Porto Nova and Santiago areas: “The Earl of Northesk, Capt. of his Majesty’s ship the Loo, being on a cruize off of Cape Finisterre, and the parts adjacent, received intelligence of a small Privateer being at Porto Nova, upon which he stood in there on the 30th of June, but the Privateer discovering him, got higher up the river than the Loo could venture, and it falling calm, Ld. Northesk was obliged to anchor close by the towns of Porto Nova, and St. Iago, into which he fired a few shot, then landed some men and dismounted 4 guns which were on a bat- tery at Porto Nova, and set fire to several houses at St. Iago. On July 7, Lord Northesk met with his Majesty’s Ship the Dealcastle, commanded by Capt. Elton, and receiving intelligence of some vessels being at Vigo, they run up the river and anchored before that town, where they made prizes of 4 vessels, 2 of which they set on fire, being light, and not having Sails on board to bring them out. They fired several shot into the Town to cover the boats while they cut away the vessels, there being a pretty smart fire at them with small arms from the shore. On July 19, upon intelligence that the privateer was still about the river of Porto Nova, the Loo run in and anchored under the Island of Blydones, where Lord Northesk put a Lieutenant and 60 men, with 2 of the ship’s 6 Pounders, into a Sloop taken at Vigo, and sent her up the river in quest of the privateer ; the Sloop could see nothing of her, but in her return chased a bark on shore, and set her on fire; and Lord Northesk landed some men, and burnt a village of about 40 houses.” 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, I3I1 Following this cruise, the Loo was again in the Channel service until May 1743, when she was ordered “To cruize between Bilbao and St. Jean de Luz,*® to intercept some Caracca ships *7 expected at St. Sebastian.” 78 At the conclusion of this cruise, which lasted some six weeks, the Loo returned to Portsmouth to refit for her ill-fated voyage to North America. Meanwhile Capt. Ashby Utting had as- sumed command under a commission dated 4 April, 1743.°° As the war between Great Britain and Spain had progressed, the people of the infant colonies of Georgia and South Carolina had felt increasing fears of an invasion by the Spanish from Florida and Cuba. In 1742 the Spanish had attacked Fort Frederica in Georgia but had been re- pulsed by the troops of General Oglethorpe at the Battle of Bloody Marsh. After this attempt by the Spanish the colonists felt that the threat of devastation of their homes and farms was even greater. Con- sequently, the Lords of the Plantations in London had been petitioned by the Governor of South Carolina to send a large warship to the Carolinas for the protection of the coastline. The result was the order- ing of the Loo to the Charleston station. Three days after Secretary Corbett signed the Loo’s orders Cap- tain Utting had received them and replied that he would “punctually comply” with them and use his “utmost endeavours” to get his ship ready for sea.*° The Loo sailed soon after Utting’s letter was posted and arrived at Portsmouth on the morning of June 18. Reporting his arrival there to the Secretary of the Admiralty, Utting complained that the 6-pounders ** on the upper deck of the Loo were “‘very indif- ferent and not fitt for a forrain voyage, being much honey combed.” *? a fact made known to him by his gunner, Samuel Kirk.** Utting recommended that a battery of 9-pounders, which had been mounted for the Hunnington, be substituted for the Loo’s worn-out 6-pounders.** 26 On the northern coast of Spain. 27 That is, ships of the Caracas (Venezuela) Company, 28 ADM 8/23. 29 Commission and Warrant Book, 1743-1745, AD 6/16, p. 335. 80 Admiralty In-Letters, ADM 1, vol. 2625, pt. 3, No. 146. 81 Heavy guns of this period were rated by the weight of the solid shot they threw. The barrel of a long 6-pounder of this period weighed around 2,000 pounds. 82 That is, the barrels, which were cast iron, had small cracks in their bores. 88 Kirk’s name is mentioned in the record of the Court Martial of Captain Utting held May 3, 1744 (Admiralty In-Letters, ADM 1, vol. 5283). 84 ADM 1, No. 417. Utting’s recommendations were not followed. This is proved by the finding of the same 6-pounders on the wreck site of the Loo. They NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 9 Preparations for the cruise proceeded swiftly. On June 20, the day after Utting wrote his letter concerning the guns, the Admiralty ordered the Captain to “make out” his pay books “to the 30 June, 1742.” *° Five days later admiralty orders “about carrying candles up and down the ship and drawing off spiritous liquors and an order to cause the men’s allowance of rum to be diluted with water when in the West Indies” were issued.*® The order directed that “whenever the ship’s Company under your command are served with Rum, Brandy, or any other spirituous liquor, instead of Beer, the same be constantly issued out to them by the Purser upon the open Deck, and nowhere else ; and that you do order all officers and others under your command, never to draw off any arrack,*7 rum, brandy, or other spirituous liquors in places under deck, but always upon open deck.” *° bore the crowned rose, a device placed on royal guns during the reigns of the Tudors and the Stuarts. The Loo’s 6-pounders were therefore cast before the death of Queen Anne in 1714. Thus they would have been at least 30 years old at the time of the loss of the Loo—a fact borne out by Utting’s statement on their condition. 35 ADM 2, vol. 60, p. 34. A year’s delay in paying the men was a common (even usual) occurrence at this time. 86 Tbid., p. 41. These orders stemmed from the loss of the Tilbury, 60 guns, in the West Indies through fire on September 21, 1742. The incident was reported by Adm. Edward Vernon in a letter to Thomas Corbett written on the flagship Boyne in Port Royal harbor, Jamaica, October 3, 1742 (Admiralty In-Letters, ADM 1, vol. 233, extracts.) “I am heartily concerned for the melancolly account lately brought me by Captain Lawrence late of the Tilbury, who came in here the 24 September in the Island Sloop, with part of his officers and men, another part remaining on board the Defyance, in execution of my orders, and upwards of a hundred of them having perished in the sea or fyre, on her acci- dentally taking fyre, and burning, and sinking in the sea, amongst which are the Master, Boatswain, and Gunner, and a Marine Officer. But I cant proceed to enquire in it at a Court Martial, til the return of the Defyance, many evidences that saw the first of it, being absent in the Defyance, so all I can say of it at present is, that it took its rise from a Marine soldier’s snatching to get a bottle rum, out of the Purser’s boys hand, who had a candle in the other hand, declar- ing he would have a dram, and in the struggle with the boy, the bottle falling and breaking, and the candle with it the rum took fire, and communicating to more in the Pursers cabbin where the fyre first began, that could not be extin- guished by all their diligence afterwards, tho they say, they threw all their powder into the sea.” Admiral Vernon at the same time submitted a copy of a general order he had published to his forces two years before requiring that the rum ration be served to the men on deck, and that it be diluted with water. The new concoction became known as “grog” after Admiral Vernon who was called “Old Grog” from his habit of wearing a “grogram” cloak. “Grogram” was a coarse material of silk and mohair. The name is derived from “gros-grain.” 87 A drink distilled from rum. 88 Admiralty Out-Letters, ADM 2, vol. 59, p. 380. IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 Another order dated the same day directed Utting to have his ship “vichialled to four months only *® for a forreign voyage, and what beer she cannot take in to be made up with good brandy . . . and to cause half of one and half of the other to be issued.” It also instructed Utting to load the food as quickly as possible and “to take care, that the brandy supplied . . . be good and wholesome,” and to report to the Lords of the Admiralty “the usefullness of the allowance of half brandy and half beer and what effect it has upon the health of the men.” *° Utting silently conformed with these orders as they were received, but on July 2, still hoping to receive the battery of g-pounders to replace his wornout 6’s, he wrote: “The time for taken in my guns draws very near, and ye officers of ye ordinance here has no orders concerning ye 9 pounders, which I had wrote for and was in hopes I should have had them as ye ship would well bare them and make her a much better man of war.” ** He also requested that, if possible, he be told his ultimate destination since he knew only that he was to go to North America. On July 11 orders were issued to the commanding officers of the Rye, 20 guns; Flamborough, 20 guns, and the sloop Spy, 8 carriage guns and 12 swivels, all stationed in South Carolina, to place them- selves under the command of Captain Utting upon his arrival there. Utting was to carry these orders with him.*? The next day the Ad- miralty issued instructions to Utting concerning the impressment of seamen while in America, furnishing him with three press warrants. The instructions cautioned him that “it is not meant, that the trade of His Maj’s. subjects in America, or ships provided with Letters of Marque to cruise against the Enemy should be distressed thereby, but only that such prudent use be made of the said press warrants as may enable you to procure men to make up your complement, when proper opportunities offer it. You are to take great care, that no indiscreet or unreasonable use be made of them.” The instructions also directed that Utting was “never to molest the chief officers, such as the master, mate, boatswain or carpenter, or any seaman found on board with protections granted by us, pursuant to Act of Parlia- mient., 89 Thus rescinding the order of June 14, which had directed the loading of a 6-months’ supply of food. 40 Admiralty Out-Letters, ADM 2, vol. 60, p. 42. Beer had been a standard beverage in the English Navy since earliest times. Easy to keep, it was superior to water, which grew putrid in the casks. 41 ADM 1, vol. 2625, pt. 3, No. 418. 42 ADM 2, vol. 60, p. 79. 43 Thid., p. 80. NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON II The same day detailed orders covering all phases of the cruise to North America were issued. They are an excellent example of the type of orders of that period given to senior officers destined for inde- pendent duty in remote parts of the Empire, and they are here quoted in their entirety.** TO CARRY GOVERNOR CLINTON TO NEW YORK AND THEN ATTEND ON SO. CAROLINA Whereas we have appointed His Maj’s. ship under your command to carry the Hon. Geo. Clinton, Esq. to his Government at New York, and then to attend on the Colony of South Carolina, you are hereby required and directd to make all possible dispatch in getting her compleated in all respects for the Sea, and you are to receive on board the said Mr. Clinton, with his Family and Equipage, and give them passage to New York, vichialling them as your Ship’s Company during their continuance on board, and allowing the Governor all such accom- modation as the Ship will afford. And whereas the ship under your command is only ordered to be vichialled to four months, and to have two months French Brandy instead of two months beer; and the Comrs. of the Vichialling having a large quantity of Brandy in store at Guernsey, in the Charge of Mr. Nich S. Dobree, a merchant in that Is- land, you are in your way down the Channel, to call off of Guernsey, without going into the Port, and send the enclosed letter with your Purser on Shore to the said Mr. Dobree; and receive from him such a quantity of Brandy as you think necessary for the use of your Ship’s Company and you can conveniently receive on board, which when you have done, you are to proceed directly to New York, without touching at the Madeiras, and there land the Governor, with his Family and Equipage; and having so done, you are to proceed on to South Carolina. And whereas His Maj’s Ships the Rye, Flamborough, and Spy Sloop, are stationed at South Carolina, You are to take them under your command, their Captains being directed to follow and observe your orders. When you arrive at South Carolina, you are to communicate these our instruc- tions to the Governor and Council of that Province, and to consult and advise with them from time to time, in what manner the ships under your command may be best employed in guarding the coasts, and securing the trade of that colony from any attempts of the Enemy, and to govern yourself according as shall be agreed on, using your best endeavors to take or destroy all such ships or vessels of the enemy, as shall come upon the coasts of the said Colony. And whereas it has been represented to us, that the Coast of North Carolina is very much infested with Spanish Privateers, who have even landed in the Country and carried off hogs and black cattle, to the great terror of the inhabi- tants of those parts, you are, when you see proper occasions, to extend your cruize as far as Cape Hatteras, or to order one of the ships under your Com- mand to do so, for the better protection of the trade of His Maj’s. Subjects in those parts; and you are to acquaint the Governor of North Carolina with this part of our instructions. 44 Admiralty Out-Letters, ADM 2, vol. 60, pp. 81-84. I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 And whereas it has been represented to us, that the Town of St. Augustine depends much upon what comes by Sea for provisions, and would be greatly distressed, if His Maj’s. ships stationed at Carolina would sometimes cruize off that Port, to prevent provisions being carried to that place by Sea, you are to have a particular regard to that service, as far as may be consistent with the other necessary services on which you may be employed. And whereas we have directed the Captains of His Maj’s. ships attending on Virginia to hold a constant correspondence with you, you are, whenever you shall find the Enemy’s ships to be too strong for you, to send immediate advice thereof to the Captains of the said ships, whom we have directed to repair to your assistance, and you are to endeavour joyntly to take or destroy them. And if the Captains of the said Ships shall at any time send you notice of the Enemy being too strong for them, you are with all possible diligence to proceed to their assistance, communicating in the first place the intelligence you have received to the Governor and Council of South Carolina, and receiving their concurrence for your so doing, and when the service is performed, you are to return to your station. And whereas the Captains of His Maj’s. ships stationed in America, have of late years taken a very unwarrantable Liberty of lying in Port with their Ships, for the greatest part of the time they have remained abroad, to the dishonour of His Maj’s. service, and the disservice of the Colonies for whose protection they are appointed, and we being determined not to suffer any such neglect for the future, do hereby strictly charge and direct you to keep constantly at Sea, when the weather will permit, and cruize in proper stations for meeting with the Enemies ships or privateers, and for protecting the trade of His Maj’s. subjects, and guarding the said colony of Carolina from any attempts of the Enemy. You are not to fail to transmit to us, one in every two months an exact copy of the Journal, that it may be seen what care and diligence you have used in putting our instructions in execution and to order the Captains of His Maj’s ships under your command to do the same. And in order to enable you the better to keep the ships under your command in a good condition to cruize and protect the trade, as well as to annoy the Enemy, you are to cause them to be cleaned once in six months, at such times as it can be most conveniently done. When the ships you command are in want of provisions, you are to apply to the Contractors of the vichialling at Carolina, for the same, and never to leave the said Colony defenceless by going somewhere to vichial; and you are to take on board no more provisions at a time, than are necessary for the service on which you are employed. You are not to hoist the Union Flag on board the Ship you Command, on account of the Governor’s being on board, or on any other pretence whatever. In case of the death of any of the officers of the ships under your command, you are to appoint such other persons to act in their names, as by the quality of their Employments ought to succeed therein. When you shall receive our orders to return to Great Britain, you are to take in no more provisions than shall be sufficient to compleat what you may have on board to three months of all species at whole allowance, upon the penalty of making good what damage, His Majesty may otherwise receive thereby. You are, as you pass through the channel, to examine such ships and vessels as you shall meet with passing from Great Britain or Ireland to France, which NO. 2 LAST CRUISE OF H.M.S. ‘“LOO”—PETERSON 13 you shall reasonably suspect to have Wool 4 on board, and upon discovering any with that comodity in them, to send them into the nearest Port, and deliver them into the care of the Collector of the Customs, in order to their being prose- cuted according to Law. You are by all opportunities to transmit to our Secretary for our information, an account of your proceedings, and of the condition of the ships under your command as to the number of men, and all other particulars and in case of in- ability by sickness or otherwise, to be careful to leave these our instructions with the next Commanding Officer. Given 12th July, 1743. Wo ifcCr0.Giv By By TG Capt. Utting, Loo, Spithead. Captain Utting must have received oral instructions that he was to carry Governor Clinton to New York several days before he re- ceived the above orders. In fact the Governor had either visited the ship or had otherwise instructed Captain Utting on the accommoda- tions that he desired aboard the Loo. Five days before the detailed orders on the cruise were written Utting had written Corbett “the carpenter will have compleated every conveniency Mr. Clinton desires by tomorrow night . . .” * On July 14 Utting acknowledged receipt of the orders of July 12 and reported that his ship was “in all respects fitt for sea.” 47 Four days later the Admiralty instructed Utting, who was now at Spithead ready to sail, to convoy the storeship Pegasus “laden with naval stores for New York and South Carolina” to America, ordering that he “convoy her safely to New York, where you are to cause her to be unloaden as soon as possible, and then proceed with her to South Carolina.” 4 At the same time additional instructions on cruising while in America were issued.* TO CRUIZE BETWEEN CAPE FLORIDA, AND THE NORTH WEST PART OF THE GRAND BAHAMA WHEN THE SEASON OF THE YEAR WILL NOT PERMIT HIS CRUIZING OFF CAROLINA. In addition to our instructions to you dated the 12th instant, you are hereby required and directed, when the Season of the Year is not proper for your cruizing on the Coasts of South Carolina, and that neither the said Colony, nor 45 The export of English wool was absolutely prohibited at this time in an effort to encourage the English woolen industry. The demand for English raw wool in the lowlands was great, and consequently the smuggling of it to the con- tinent was profitable. 46 ADM 1, vol. 2625, pt. 3, No. 419. 47 Thid., No. 420. 48 ADM 2, vol. 60, p. 96. 49 Thid., pp. 96-97. 14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I that of Georgia is under any apprehension of being molested by the Enemy from Havanna or Augustin, to proceed with His Maj’s. Ship under your Command and Cruize between Cape Florida and the North West part of the Grand Ba- hama, ’til such time as the Season will permit your return to Carolina, taking care to have a sufficient quantity of provisions on board to last you on that service. You are diligently to look out for the Enemy’s ships passing through the Gulph of Florida for Europe, and use your utmost endeavours, to take, sink, burn or destroy them. But before you proceed on this Service, you are to communicate your design to the Governor of Carolina, and not to go thereupon, if you find any reasonable objections thereto. Given 18th July, 1743. Wak. sB. By TC Capt. Utting, Loo, Spithead. Utting had been thinking of the same operations plan as that of the Lords of the Admiralty, for on July 19, a day or so before he received the additional instructions, he had written: I was a little hurried to save post with my last yet dont know whether I ex- plained ye plans and time of cruising so plain as you could wish for fear of which beg you'll be pleased to indulge me with this to acquaint you. I propose (if ye service will allow me and you can git me orders) to saile from South Carolina ye 10 or 15 of October and cruise in and about ye Gulfe of Florida, as far as ye Cape °° if I can git there till ye middle of Jany. then return to Caro- lina. And as soon as I can water, victuall, and refitt, in all respects, then pro- pose to saile, and cruize on ye coast of Carolina of [off] St. Augustine or on such part of ye coast as I shall find the service require me most. Given ye 20 gun ships proper stations as ye service shall require, on this coast I propose to keep all ye summer months 51; ye latter end of May shall go in for 6 or 8 days to victuall and water and then cruize till ye 20 or 25th. of July when as I shall be then about 12 months foull shall go in to heave down and about ye 2oth of Septr. shall saile to cruize on ye aforesaid station: yet I never propose to be in port above 2 months in ye year; after my first careening shall heave down every 6 months. But as I am graved and tallowed 52 can go 12 months at first. There is an exceeding good careening place at Port Royall 5% which can be made ours 50 Cape Florida. 51 That is, keep to the sea during the summer months. 52 See footnote 5, p. 5. In navy yards graving was usually done in a drydock. On remote stations it was necessary to careen the vessel by mooring her in a river, unloading her, and then “heaving her down” by pulling her over with tackles secured to trees on the bank. In this position half of her bottom was above water and could be cleaned. The process was repeated for the other side of the bottom. Hulls were coated with tallow as a protection against growths and water penetration of the ship’s planking. 53 South Carolina. NO. 2 LAST CRUISE OF H.M.S. ““LOO’’—PETERSON 15 conveniently to heave down without expense to the government. I have wrote to ye Navy Board for careening gear, but have not had an answer.®4 On July 25 the Captain acknowledged receipt of the further instruc- tions on cruising and the orders to escort the Pegasus, and prepared to set sail.°° On August 6 Governor Clinton, his wife and her children, and suite of 15 persons came aboard the Loo.*® She probably sailed within a week, Six weeks later the Loo arrived safely in New York harbor with her charges, and the Pegasus. The Governor and his suite disem- barked on September 22.°7 Utting reported in a letter dated in New York Harbor September 29 that the voyage had been uneventful “with nothing worth their Lordships notice.” In the same letter the Captain made his first report on the trial ration of half brandy and half beer, stating that it agreed with men “extreamly well, and they are well pleased.” °* The ship, he reported, was unmooring as he wrote, and expected to sail that afternoon for South Carolina escorting the Pegasus. His departure was delayed until October 6, however, probably by adverse weather, but the bright lights of New York might have been the real reason, since Utting mentioned no cause for the delay. After a passage of five days the Loo arrived off Charleston Bar. In Charleston he found the sloop Spy ready for sea, the Rye “cleaned and almost fitt for sea,” the Flamborough “sheating.” °° He immediately delivered the Admiralty orders instructing the captains of these ships to place themselves under his command, and then issued orders giving each ship stations for cruising off the Carolina coast for the defense of the colonies and protection of English and colonial ship- ping. Captain Hardy of the Rye was directed to “cruize on the coast of South Carolina, between Charles Town Barr and the So.W most part of the same coast, keeping off St. Augustine, and as near into the shore as you shall judge proper when winds and weather will permitt to intercept any trade that may come from the Havanah to that place.” °° Hardy was also instructed to inform the Governor of 54 ADM 1, vol. 2625, pt. 3, No. 421. 55 Thid. 56 ..o0’s General Muster Book, ADM 36, Ser. I, vol. 1823. 57 Loo’s General Muster Book, ADM 30, Ser. I, vol. 1823. 58 ADM 1, vol. 2625, pt. 3, No. 423. 59 Tbid., No. 435. British ships were at this time sheathed with thin fir boards backed with horsehair and pitch. The sea worms ate through the thin board but were repulsed by the hair, and the ship’s hull planking was thus protected. 60 ADM 1, vol. 60, No. 435. 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 Georgia °*' of his activity off that coast, to remain at sea as long as his water and provisions would permit, and, after returning to Charles- ton to take on supplies, to return to his station and cruise as soon as his ship was ready for sea. Captain Hamar of the Flamborough was directed to cruise off the coast of North Carolina. “Whereas the Rt. Hono. the Lords Commiss. of the Admiralty has been informed that the coast of North Carolina has been much infested with priva- teers [Spanish] to the great determent of the inhabitants of that province, you,are to use your utmost endeavours to take or otherwise destroy them or any of the enemy’s ships as you may possible meet with in your cruise.’ °* Hamar was directed to inform the Governor of North Carolina ** of his cruising on that coast and told to keep to sea at all times possible.® Captain Newman (also spelled Newnham) of the sloop Spy was ordered to join the Loo and cruise with her until further orders.®® The execution of Utting’s plans was to be delayed, however, for the next day an “exstream hard gale of wind at ENE” struck the Loo as she lay at anchor off Charlestown Bar, obliging Utting to cut his “best bower cable’ ®*’ and go to sea “for fear of a hurricane.” For four days the Loo rode out the gale at sea, and when the storm was over Utting returned to his anchorage off Charleston, recovered his anchor and the next day (Saturday, October 25) “‘saild for Port Royall to refitt having received great damage in . . . masts and rig- ging.°> Upon examining the damage to the Loo, Utting and his offi- cers found the main yard sprung in three places and unserviceable. “The mainmast sprung in ye lower partners ® about 6 inches in tho not to bad but shall be able to fish *° him and make as serviceable as ever...” 7 Utting was mistaken, however, in his estimate of the damage, for closer examination revealed extensive damage to the mast 61 Tbid. 62 Tbid. 63 ADM 1, vol. 60, No. 436. 64 Tbid. 65 Tbid. 66 Tbid. 67 The cable of the heaviest of the two anchors carried in the bow of a ship. The bower anchors were those used for anchoring under ordinary conditions of wind and sea. 68 ADM 1, vol. 2625, No. 438. 69 Planks fitted snugly around the base of a mast, a hatch, or a capstan cov- ering the opening in the decks. 70 To splice a broken spar or mast by binding with splints and wedging firmly. 71 ADM 1, vol. 2625, No. 438. NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 17 below decks and the Loo was not to leave Port Royal until December 30, when she began her last cruise. While lying at Port Royal Utting continued active direction of the vessels under his command from the Loo. On November 18 he issued two orders to Captain Newman of the sloop Spy. The first directed Newman to watch for a vessel expected from Havana with prisoners of war which were being exchanged and, should he meet with her, “to take out thirty of the best seamen on board for the service of his Majesty’s ship Loo.” * The second order directed the captain of the Spy to keep close touch with Charleston to obtain intelligence of the expected declaration of war against France, and if hearing of such declaration to rendezvous with the Loo.” By Captain Ashby Utting, Commander of his Majesty’s Ship Loo Whereas we are in dayly expectations to hear of the Declaration of a French War. You are hereby required and directed when on your cruise to call as often of Charles Town, as you shall think convenient to get the best information you can. And when you find any certain intelligence of a French War being declared either by Publick or private letters. You are immediately to proceed and joyn me of the N W part of the Grand Bahama, and if not find me there to proceed of the Isaack Rocks and the Bominies and if not at either of those places to proceed of Cape Florida and the Martiars [Fla Keys] 74 where you are to cruise for me Ten days and if not find me in that time you are to proceed to Hinds Bluff one of the Burry Islands where you are to fill up your water and then proceed and cruise between the N W part of the Grand Bahamas and Cape Florida till you meet me or as long as your provisions will last; and then return to Port Royall where you are to compleat your water and provisions to three months and to proceed to sea, and cruise between that Port and Georgia till further orders. Given under my hand on board the said ship in Port Royall Harbour the 18th day of November 1743. Ashby Utting To Captain Newnham of his Majesties Sloop Spy. 72 ADM 1, vol. 2625, No. 426. 73 ADM 2, vol. 2625, No. 426. 74 Ponce de Leon named the Florida Keys “the Martyrs” because, he said, from the sea they bore a resemblance to the early Christian martyrs tied up on lines of stakes for execution. From 3 miles or so at sea the larger trees on the Keys indeed appear in long rows, the low-lying land of the Keys being out of sight over the horizon. In an age of Christian fervor, when religious signifi- cance was seen in every natural phenomenon, such an analogy would be the expected thing. The name “Martyrs” appeared on charts as late as the early 1800's. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 On November 25 Captain Hamer of the Flamborough was also ordered to be on the lookout for the prisoner exchange ship expected from Havana, and to remove seamen for the Loo.*® As work proceeded on repairing the storm damage to the rigging of the Loo, the carpenter discovered that the mainmast had been sprung in several places and reported to the Captain. Utting, on November 27, appointed the first and second lieutenants, the master,” the carpenter and the carpenter’s mate to “take a strict and careful survey” of the mast and report its “exact condition” to him.’” The survey board acted immediately and reported the same day that the mast had been severely sprung below decks and in their opinion was unfit for service.”® For over a month the crew and officers turned to getting a main- mast cut and rigged, during which operations Warren Bolitha, the First Lieutenant of the Loo, broke three ribs, and on December 29, the day before the Loo sailed, he requested the captain to “let him go home in order to get cured.” *° While Utting was struggling to refit the Loo’s damaged rigging, a letter arrived on December 14 from Capt. Charles Hardy of the Rye announcing that she, too, had sprung her mainmast. Utting immedi- ately ordered Hardy to replace the mast “as soon as possible” and to return to his station off the Carolina coast. As a precaution against confusion on the part of the commanding officer of any ship that might relieve the Rye while the Loo was away on its expected cruise toward Cuba, Utting instructed Hardy to pass on his orders to his relief.*° Meanwhile the Flamborough had been at sea and had fallen in with the ship that was bringing freed prisoners exchanged in Havana.** On December 15 Utting ordered Captain Hamer to search out the Spy, which was then cruising off Charlestown Bar, and transfer to her, for transportation to the Loo at Port Royal, 30 of the seamen whom he had impressed, and then to proceed to cruise off Georgia, sending a boat to the Governor of that colony “for any intelligence he may have of any of the enemy’s ships, or vessells being on that coast.” 8? 75 ADM 1, vol. 2625, No. 432. 78 Warships of this period had an officer in charge of the active sailing of the ship known as the Master. 77 ADM 1, vol. 2625, No. 433. 78 Thid. 79 ADM ft, vol. 2625, No. 431. 80 ADM 1, vol. 2625, No. 430. 81 Among them, John Manley and Henry Spencer, who were to play a fateful part in the subsequent events. (ADM 1, vol. 2625, No. 446.) 82 ADM 1, vol. 2625, No. 434. NO. 2 LAST CRUISE OF H.M.S. ‘‘LOO”—PETERSON 19 At the same time he ordered the Spy to take aboard the 30 seamen from the Flamborough and then to cruise off Port Royal Bar and join the Loo when she came out.** On December 22, Utting ordered Captain Ward of the ship Tartar, which had arrived to relieve the Rye,** to cruise on the Carolina and Georgia coasts on the Rye’s old station.*® Finally, on December 30, work on the Loo having been completed and the winds and tide favorable, the ship crossed the bar at Port Royal and began her last cruise. In a final letter to the Admiralty before the ship weighed anchor, Utting explained the long delay occa- sioned by damage the Loo had received in the storm off Charleston October 16-20, which he had underestimated in his letter to the Ad- miralty dated November 12, 1743, at Port Royal Harbor. He re- ported that it had taken him more than a month to get a new mast cut, partially seasoned, and rigged ®* and took occasion to point out again to the Lords of the Admiralty the desirability of cutting several trees and seasoning them as a reserve to be used for the manufacture of masts or yards in the event of further damage to the ships under his command. At the same time Utting reported that he had relieved his first lieutenant, Mr. Bolitha, because of his injury, so that he could return home to England, and had promoted his second lieutenant and third lieutenant each one grade, then filling the vacancy left by the third lieutenant by the appointment of one William Lloyd whom he de- scribed as “a young gentm. well qualified for Preferement in his Maj’s. service.” 87 After this last word from Utting, the Loo sailed to her station in the Florida Straits and began cruising against Spanish shipping. The morning of Saturday, February 4, 1744, found her cruising in the Straits off Havana. Around 8 o’clock in the morning a sail was sighted, and the Loo gave chase. As the stranger was neared, two seamen of the Loo, John Manley and Henry Spencer, who had been in the group of prisoners exchanged from Havana, informed Utting that they recognized the ship as the Billander Betty on which they had served. They told Captain Utting that while on a voyage in the 88 ADM 1, vol. 2625, No. 427. 84 Captain Newman of the Rye had been directed to convoy merchant ships to England from Charlestown in an order dated September 23, which was sent out by the Tartar. (ADM 2, vol. 60, p. 270.) 85 ADM 1, vol. 2625, No. 429. 86 ADM 1, vol. 2625, No. 424. 87 ADM 1, vol. 2625, No. 424. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, I3I1 Betty (Capt. John Eades) from England to the Isle of May ** and South Carolina they had been captured by a Spanish vessel off the coast of South Carolina. The Spanish put aboard a prize crew and sent the Betty on to Havana, but the Spanish vessel herself was lost in returning to Havana. While prisoners in Havana, Manley and Spencer had heard that the Betty had been converted to a “snow” and was to make a voyage to Campeche. About noon, when the Loo came alongside her chase, Utting sent an officer to examine the stranger’s papers. Upon hearing that the master of the quarry could show only a common receipt, Utting decided to seize the ship for the proprietors of South Carolina and send her to Charleston. Before sending her off, however, Utting requested that an “Trish gentleman” on the snow be brought aboard the Loo for questioning. Before this gentleman left the snow, he was seen to throw a large oil- skin packet overboard. A boat from the Loo retrieved the packet and Utting discovered that it contained papers in French and Spanish. At this, he decided to take the prize in, with the Loo as escort. The examination of the prize had taken the entire afternoon, and when the Zoo set sail with her charge it was 6 p.m. and growing dark. The “Pan of Matanzas” (fig. 1), a flat-topped mountain behind Matanzas Bay on the coast of Cuba, bore south by east at a distance of 18 to 21 miles.*° Taking his fix on the mountain, Utting set sail and ordered a course northeast by north, the wind coming from the southeast. This course was kept until midnight, when Utting, be- lieving that he was clear of the Double Headed Shot Key in the west- ern end of the Salt Key Bank, instructed Randell, the first lieutenant and officer of the off-going watch, to alter the course to northeast and went below to his cabin to rest, having been continuously on deck since early morning.°° Shortly after, Randell was relieved of the watch by Robert Bishop, the master. Randell relayed these instructions to Bishop, reminding him to have the deep-sea lead line cast every half hour,®t and went 88 “Maio” in the Cape Verde group occupied until the end of the eighteenth century by the English, who claimed a right to the island under the marriage treaty between Charles II and Catherine of Braganza of Portugal. The English occupation is recalled in the name “English Road,” which the port of Nossa Senhora de Luz is sometimes called. 89 The bearing and the distance to the Pan of Matanzas were given by Lt. James Randell in his deposition to the court martial that tried Captain Utting. (ADM 1, vol. 5283.) 90 Utting’s letter of February 15, 1744. (ADM 1, vol. 2625. (See Appendix B.)) 91 Bishop’s deposition at the court martial. (ADM 1, vol. 5283.) NO, 2 LAST CRUISE OF H.M.S. ‘‘LOO’”—PETERSON 21 below. Nothing to arouse the suspicion of Utting or Randell had been seen during the latter’s watch. The night was dark and cloudy, with visibility not over a quarter of a mile.*” At 12:30 a.m. and again at 1:00 the deep-sea lead line was cast according to orders, and no bottom was found at 300 feet. At about 1:15 Bishop sent the lead-line crew to the side to clear the line for heaving and followed them to the gunwhale himself to see to this. To his great surprise he found the ship in “white water” and saw breakers ahead. He instantly “ordered the helm alee” and sent a message down to Captain Utting ** to call him on deck. As Utting rushed on deck he found the ship coming about into the wind and away from the reef on which the breakers were rolling. As the ship veered off the wind the head sails were caught across wind and the ship struck the reef Lh bast . Uw Pan de VMataneas Twat Matiigns Fic. 1.—The Pan of Matanzas from a vignette appearing on an English chart dated 1794. aft.°* At this the mainsail was set “in order to press her off,” and Utting ordered a boat out to sound around the ship.** The officers and men off watch and sleeping below were awakened by the shock of the ship striking the reef. John Vivian, the carpenter, rushed aft, whence the shock had come, and found the tiller broken off. He re- ported this to Utting just as another swell caught the ship and broke - off the rudder, at which she began shipping water in the hold. Utting ordered all pumps manned, and the water in the hold began to fall, but as the crew was getting out the boats “three or four severe seas” crushed the ship against the reef and she began sinking rapidly. When it became apparent that the ship could not be saved, Captain Utting ordered Mr. Bishop and Gunner Samuel Kirk to save as much 92 Deposition of John Randolph, master’s mate, at the court martial of Cap- tain Utting. (ADM 1, vol. 5283.) 93 Bishop’s deposition. (ADM 1, vol. 5283.) 94 Utting’s letter of February 15, 1744. 95 Bishop’s deposition. (ADM t. vol. 5283.) 96 Utting’s letter of February 15, 1744. 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 of the bread and gunpowder as possible before the water rising in the hold ruined it. Bishop was able to save 20 bags of bread before the water forced him from the breadroom, and by the efforts of Gunner Kirk 6 barrels of gunpowder were saved. The prize, which had struck the reef just after the Loo, was rolling and pounding herself to pieces on the coast and, in order to save the men aboard her, Utting ordered her masts cut away and her guns and 3 —— HAMAS ABA Cb Fic. 2—Map of the Florida Straits, showing the course of the last cruise of the Loo. 1, Havana. 2, Pan of Matanzas. 3, Double Headed Shot Key. Dotted line, Captain Utting’s intended course. Solid line, actual course of the Loo. anchors thrown overboard. After this she lay more quietly in the swell and her men were saved. With the coming of daylight Utting and his officers saw with great surprise that they were ashore on a “small sandy Key about 13 cables *? length long and 4 broad which lay on the edge of the Bank of the Martiers 3 leagues °* without them.” *° (See fig. 2.) At no time since the ship had struck had the pilots or officers 97 A cable was 600 feet. 98 A league was 3 nautical miles. 89 Utting’s letter of February 15, 1744. NO. 2 LAST CRUISE OF H.M.S. “LOO’—PETERSON 23 doubted that they were aground on Double Head Shot since, under normal circumstances, the course that they had steered could not have carried them to the Florida Keys. Utting sent Lieutenant Randell ashore to see if fresh water was to be had there, but he found none.*°° As full daylight came Utting landed all the men from the Loo and the prize with the exception of a few who were employed in cutting holes in the frigate’s deck to recover casks of water and such other supplies as could be saved. At about Io o’clock, to Utting’s great joy, a sloop was sighted offshore and a signal was made to her, but the sloop stood out to the northwestward.’ The captain immediately armed all the boats and with Lieutenant Randell and Mr. Bishop in command sent them in pursuit, instructing them to exert every effort to bring the sloop in, since it probably would be their only chance of succor. The desperate situation of the group was evident to all. Here were some 280 men stranded on a small sandy islet just off a hostile coast swarming with the savage Caloosa Indians who murdered Englishmen on sight.*°? To add to the insecurity was the evident fact that in a blow of any force the whole islet would be swept by waves. At night Utting posted watches, each consisting of 25 marines and 25 sailors, around the island at the water’s edge as “centenells” to prevent a surprise night attack from the Caloosas “the Indians hav- ing numbers of canoes.” 1° The next morning (Monday) as daylight came, Utting and the men ashore were overjoyed to see the boats bringing in the sloop. As they came ashore Randell and Bishop reported that on the approach of the armed boats the Spanish crew had abandoned the sloop and were no doubt now headed for Havana in their boat. Meanwhile the men, frightened and confused, became “very rebel- lious and mutinous dividing into parties and growling amongst them- selves,’ *°* claiming that the officers no longer had authority over them, and clamored to leave the island immediately. Utting took no notice of them but, with the men who would work, continued efforts to recover water and other provisions from the wreck. All day Tuesday was spent in getting water casks from the Loo’s hold and in getting the sloop and boats ready for the escape. The 100 Randell’s deposition. 101 Tbid, 102 Spaniards fared a little better, as the Caloosas knew they could be ran- somed. 103 Utting’s letter of February 15, 1744. 104 Tbid, 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 capacity and safety of the longboat were increased by adding planks to the gunwhales, giving her a higher freeboard and decreasing the chances of shipping water in a seaway while heavily loaded. Thus altered, the boat was able to carry 60 men.*” At about noon the next day, Wednesday, February 8, all the men were embarked—6o in the altered longboat, 10 in the yawl,'°* 184 “in the little Sloop not 30 tuns’’ *°7 and 20 in the captain’s barge. Utting sent the sloop, the longboat, and the yawl 3 or 4 miles offshore while he remained behind with the men detailed to the barge and laid most of the gunpowder which had been saved and some other fuel along the starboard gun deck of the Loo, the only deck remaining above water. By 2 o’clock all the preparations were completed and Utting fired the ship. As they rowed away the ship blazed to the top of her masts and blew up “in several places and was in flames fore and aft.” 1°° The burning ship was visible until sunset and while her hull was completely destroyed, Utting feared that the Spaniards would return and salvage her guns and anchors, since intelligence of the Loo’s end would be communicated to the Spanish in Havana by the escaped crew of the sloop. Utting’s plight was still grave, since the sloop was very much over- loaded and would have capsized in a blow. He placed First Lieutenant Randell and Third Lieutenant Lloyd in charge of the longboat, his Sailing Master Bishop was given command of the barge and “a mate” assigned to the yawl. Utting remained in command of the sloop. The motley fleet set a course for the Bahamas, the boats being ordered “in case of separation to make the best of their way over to the Bahama Bank for Providence.” *°° That night Utting carried a light to guide the boats, but they out- sailed the overloaded sloop and at midnight were lost from sight when Utting had to tack and stand to the northward after signaling his change of course with the light. At daybreak the boats were gone, and Utting, feeling that they were bound for Providence and being unable to set sail for an easterly course, set a course for South Carolina. In his report to the Admiralty he summed up the desperate situation in which he found himself with the overloaded sloop: “it blowing fresh and the sloop top heavy with men could not carry sail so obliged to 105 Tbid. 106 A heavy double-ended rowboat. 107 Utting’s letter of February 15, 1744. 108 Tbid. 109 Thid. Providence had been settled in the seventeenth century. NO. 2 LAST CRUISE OF H.M.S. ‘“LOO’’—PETERSON 25 bear away and take my fate through the Gulph of Florida **® for any port of Carolina even for St. Augustine (if I could fetch nowhere else) rather than all be drowned which Doe assure you had very little other prospect.” +14 The fair weather continued and the overloaded sloop arrived in Port Royal harbor (pl. 17) on the night of February 13. Utting and the men were worn out from physical and mental strain, all realiz- ing that their escape from capture or drowning was just short of miraculous. Upon his arrival at Port Royal Utting began immediate steps to assemble evidence to protect himself in the court martial that he had to face for the loss of the Loo. His first step was to send one of his pilots, William Lyford, to the town of Beaufort 6 miles north of Port Royal to give a deposition before Robert Thorpe, justice of the peace. In the deposition Lyford stated that in his opinion the course the Loo had steered before she ran aground “was the best through the Gulph (and is generally allowed so to be) and was then of the opinion that such course would carry the said ship nearer the Bahama shore than the Florida; and this deponent further deposith and makes oath, that he is well acquainted with the Gulph of Florida having used it these thirty years past.” 12? Eight days later, on February 21, Utting was in Charleston start- ing proceedings to prove that the prize which he had taken was a legal one. John Manley and Henry Spencer, the two seamen who had recognized the prize as their former ship, appeared before James Grome, judge of the Court of Vice Admiralty of the Province of South Carolina, and swore under oath that the prize was the former Billander Betty, and that while on a voyage from England to the Isle of May and South Carolina, it had been captured off the coast of South Carolina on April 9, 1743— by a Spanish vessell bound from the Havannah to St. Augustine with about sixty or more soldiers on board, that the said vessell not being able to make St. Augustine return’d to the Havannah and in her passage was cast away, that the Billander so taken as aforesaid was carried to the Havannah and was there converted into a Snow and intended on a voyage to Campeachee but afterwards these Deponents hear’d that she was bound for the Mississippi. That these Deponents came to this province with the Flag of France and were press’d on Board his Majestys Ship the Loo under the command of Capt. Ashby Utting, that on a cruise in the said ship they met with a Snow which these Deponents very well knew to be the Billander Betty taken as aforesaid 110 The Gulf Stream would carry him northward. 111 Utting’s letter of February 15, 1744. 112 Deposition accompanying Captain Utting’s letter of February 15, 1744. 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I by the Spaniards and converted into a Snow that the said Snow was taken by the Loo about a fourthnight ago, viz the 5th of this instant February and cast away with the said Man of War about nine leagues to the westward of Cape Florida.118 On March 6, as the hearing proceeded, this deposition was intro- duced to the Court of Vice Admiralty as “Exhibit Marked I” and undoubtedly had a great influence in the outcome of the case in Utting’s favor. By March 12 the proceedings in the Vice Admiralty Court seem to have been concluded. Utting had only to wait for the return of his officers from the Bahamas, and then a warship for transportation to England if his court martial could not be held in America. The loss of the Loo had immediate repercussions in the colony of South Carolina, and on the same day the case in court was concluded, the Governor, James Glen, wrote a lengthy letter to the Lords of Trade in which he indicated the fears of invasion which had swept the southern frontier of the colony: After writing so lately by Captain Hardy,115 I had not so soon troubled your Lordships with another letter, but the loss of his Majestys Ship the Loo makes it my duty; this unlucky accident happened the fifth of February about one in the Morning, by her running on some rocks called the Martyres to the South West of Cape Florida. . my principal concern is to express to your Lordships how sensibly that loss affects this province. The long neglected Town of Beauford, upon the arrival of this Ship, and the assurances given that another would be sent out, began to revive, and many good houses were built, and many grants for Town Lotts were applyed for, so that I am persuaded that Town and the adjacent Country, would soon have been well settled, and consequently our Southern Frontier, where we are most vulnerable, would have been strengthened, but now I receive letters and petitions dayly from the best People in those parts, repre- senting their fears and the dangers to which they are exposed, and everything is at a stand, tho’ I have stationed one of our gallys (a very fine small vessel) there, I have likewise desired the Captains of the Man of War on this station, to keep a particular eye upon that Port, in their Cruizes along our coast.116 118 Copies of papers relating to the proceedings of the Court of Vice Admiralty sent by Captain Utting to Thomas Corbett after his return to England (ADM 1, vol. 2625, No. 455.) 114 Utting and his wife had been residents of South Carolina several years. The wait for transportation to England was probably not too burdensome to Utting. 118 Of the Rye, which had sailed for England a short time before Utting reached Port Royal. 116 Letter of James Glen dated March 12, 1743/44. Colonial Office original correspondence, CO5, vol. 370, pp. 141-142, Public Record Office, London. NO. 2 LAST CRUISE OF H.M.S. “LOO”—PETERSON 27 He then went on to point out the suitability of Port Royal as a harbor and its strategic location in relation to the Florida Straits and the Spanish trade routes : And as most of the trade and treasure of France and Spain must come through the Gulf of Florida, where can it be so properly waited for as here, where a few great ships stationed, to cruize betwixt this and Cape Florida, a very easy navi- gation, must become masters of everything.117 The Governor reported that Captain Dansant, captain of the Loo’s prize, would be sent off without being permitted to see the fortifica- tions of Charleston, and “The forty-four marines belonging to the Loo,” he stated, were being “lodged at the expense of this government and shall be well taken care of.” 11* The sailors were no doubt taken into the other ships present on the station, the Governor not mention- ing them. Governor Glen spoke a good word for Captain Utting with the Lords of Trade, describing him as ‘a Gentleman who by a long resi- dence in this Province, has established a character amongst all ranks of people here, for strict honors and veracity, as well as for his care, diligence, and knowledge, as an officer.” 7° Early in April Utting and his officers, who had arrived from the Bahamas, sailed for England and arrived there late on the night of May 24. The next day Utting reported his arrival to the Admiralty and requested an early court martial for the loss of the Loo. (See fig. 3.) Six days later, May 31, the court of 12 captains sat on board the ship-of-the-line Sandwich with Vice Admiral James Steuart presiding. After a consideration of the depositions and testimony of Utting and his officers the Court was “unanimously of the opinion that Cap". Utting and his several officers did in no wise contribute to her going ashore, but that it was owing to some unknown accident, it appearing to the Court, that the course the ship steered was a good one, and must have carried her thro the Gulph of Florida, with all safety had not some unusual current rendered the said course ineffectual.” 1*° On June 6 the Lords of the Admiralty ordered the Navy Board to procure funds from the Treasury and pay the officers and men of the Loo through the day she was lost. On August 10, 1744, the officers and men gathered on Broad Street in London and were paid the 1,510 117 Thid, 118 Thid, 119 Tbid. 120 Report of Court Martial dated June 1, 1744, Admiralty In-Letters, ADM 1, vol. 5283. cee ee ee ee. Sef wreigS - ~, * [ye icahesdlviail fie aie, ] UO l t e i Mal. shes ean dhe f3 te att at : (2, * Bare Cre > Pig) bee to hae A008 9? | Bualoseotsd: Carpenter? Pusfirr, Jone! Sf te - Mall Peon) Sowhs h Ohh mates Bary Ain Ue orth arg onTh Srnetp Ye fhyt sak F Potty) 6 Offic art Rfelated tn ost gh an ei ll Shp, Tyee nae Gnd: Gry? Emp? albert ga: feet Anre- Eon on Kou) She’ msn -aWar Shae fonatang wher fe aot af fare fiat yf cer oo Cp e721 ort Ske. Lu.) rd prsceea, 1A aoaref fon fel fee* OLED lag He apr ie ple Fase ih © Gal a1 EG orn at Poti. Dipti wats viet (ome c Eure tna Many KL sf ry Duty fo Mayz fo Goof ‘then then foyer % Sing Lert uae I Kope- thare Kors kept ro: lh ods if 5) y onlt p baste fo Ooo’ thue So® pl aoe Di ash iasbad Bebe Sf Of? *: Shp oa rho | ie x4 often ff eonr sxe Ong ete Fic. 3.— —Letter of Capt. Ashby Utting to the Secretary of the Admira Utting to athe Secretary of the Acti ed ing his arrival in England May 24, 1744, stand trial by court martial for the loss of the Loo. (Photostat feos oe Public R ecord Office, London.) 28 ae RK ® NO. 2 LAST CRUISE OF H.M.S. “LOO’—PETERSON 29 pounds 4 shillings 11 pence due them after deductions of 1,121 pounds 4 shillings 11 pence for tobacco, clothing, hospital fund, pay advances, etc. James Compton, Captain, Royal Navy, a Navy Commissioner, kept an eye on the proceedings while navy clerks Stephen Mercer and Philip Stephens and treasury clerks John Wilson and Thomas Vaughn checked the pay list and disbursed the money.1** After his acquittal Utting attended the Secretary of the Admiralty frequently while waiting for an answer to his request for another com- mand. On June 12 he discussed with the secretary the possibility of getting command of the Mary Galley and the next day advised Cor- bett that several of the men and petty officers of the Loo desired to ship with him on his next cruise. He also reminded Corbett of the desirability of his return to the Carolina station since his wife was there:*#" On July 7, 1744, a commission was issued giving Utting command of the Gosport.1** While fitting out his new ship, Utting continued to hope that he would be able to get orders to return to the Carolinas. No one knew better than he the danger of invasion to which the colony was exposed through the loss of the Loo, and he was anxious for the safety of Mrs. Utting, who was at Port Royal. His fears were multiplied when, on August Io, he received a letter from his wife, dated July 5, in which she reported that the settlements south of Charleston had been evacuated because of fear of an invasion and that she was a refugee in the provincial capital. Utting’s patience reached the breaking point as he pleaded for orders to America: “This is a very shocking affair both to her and me and beg for God’s sake you'll be so good to use your interest with Lord Winchelsea *** to git me to some part of America.” 1”° The exigencies of war, however, outweighed the personal problems of Utting, and he was ordered to the Baltic to convoy a fleet of merchantmen to Elsinore, Denmark, and Bergen, Norway. On October 13 Utting was back in England with the convoy from Bergen. The next month he took a convoy to Ostend, leaving on the 15th and returning to England on the 24th, assuming command of the Aldborough sometime between his return and November 29 under a commission dated November 7.1*° Utting’s wish to return to South 121 [00's pay list dated Aug. 10, 1744, Admiralty Ships Pay Books, Treasurers Series I, ADM 33, No. 352. 122 Admiralty In-Letters, ADM 1, vol. 2267. 123 AD 6/16, Commission and Warrant Book, 1743-1745, p. 335. 124 First Lord of the Admiralty. 125 ADM 1, vol. 2625, No. 477. 126 AD 6/16, Commission and Warrant Book, 1743-1745, p. 380. 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 Carolina was realized shortly after, when he was ordered to escort a convoy to America and assume his former command as senior officer present at Charleston. On March 26, 1745, he arrived at his old sta- tion and began the direction of naval operations off the Carolinas. The threat from the Spanish was still real, and naval patrols were necessary to prevent surprise attacks on the coastal settlements. Sickness plagued the Alborough and Utting was unable to keep the sea as he should have. Enemy privateers arrived off the coasts of Carolina and Georgia, and the captain was at his wit’s end to protect the coastal settlements from Fredrica, in Georgia, to Charleston with his little squadron. As a result some discontented merchants in Charleston complained to the Carolina proprietors that Utting was not doing his duty. These complaints seem to have been unjustified, for the Governor, Council, and several principal merchants refused to sign them. Thus misfortune harassed Utting during his last cruise until early in January 1746, when he died on board the Alborough in Rebel- lion Road, Charleston, just after returning from a patrol off the coast. On April 18, 1744, while Utting was on the high seas returning to face a court martial for the loss of the Loo, the Lords of the Ad- miralty had ordered the Navy Board to “cause a new ship of forty-four guns to be built in the room of the Looe lately lost near the Gulph of Florida.” 1**7 Today the name “Looe” is perpetuated by the sub- merged reef lying off the central Florida Keys, visited by occasional fishermen who must wonder at the strange name it bears, never dreaming that the quiet little reef was once the scene of as dramatic a story of shipwreck and rescue as can be found in the annals of the English colonies in America. 127 Admiralty Out-Letters, ADM 2, vol. 205, p. 344. APPENDIX A EXTRACTS FROM (THE NAVY, LIST ‘RELATING TO THE. “LOO” Period April 1, 1707—May 31, 1707 June 1, 1707—June 30, 1707 July 1, 1707-September 30, 1707 October 1, 1707—October 31, 1707 November 1, 1707—November 30, 1707 December 1, 1707-January 31, 1708 February 1, 1708—March 31, 1708 April 1, 1708—-April 30, 1708 May 1, 1708-July 31, 1708 August 1, 1708-February 28, 1709 March 1, 17090-July 31, 1700 August I, 1709—September 30, 1709 October 1, 1709—October 31, 1709 November 1, 1709—-November 30, 1709 December 1, 1709—December 31, 1709 January I, 1710-January 31, 1710 February 1, 1710-February 30, 1710 March 1, 1710—March 31, 1710 April 1, 1710-April 30, 1710 May I, 1710-May 31, 1710 June I, 1710—June 30, 1710 July 1, 1710-July 31, 1710 August 1, 1710-September 30, 1710 October 1, 1712—October 31, 1712 November 1, 1710-December 31, 1710 January 1, 1711—January 31, 1711 Duty “Longreach taking in Guns” “Going to Archangell” “Gon to Archangell” “Arch-Angell” “Returned with the Russia ships to Grimsby” “Sheerness land)” “Downes for Newfoundland” “for Newfoundland” “Gon to Newfoundland” “at Newfoundland” “coming convoy from Newfoundland” “Coming convoy from Lisbon but last from Newfoundland” “Sheerness—refitting” “Sailed to Join the Tilbury Etc. at Goree and when she returns to join the Gosport and Strombolo between Dover and Beachy” “Downes ordered to cruise between Dover and Beachy head” 1 “.. . Dover and Beachy head” “Cruizing between Dover and Beachy head” “Holland—ordered to cruize between Dover and Beachy Head” “Holland, to come to the Downs” “Dunkirke Squadron” “Sailed to Cruize on ye French Coast between Cape Barfleur and Harve de Grace” “Dunkirke Squadron” “Margate Roads ord. to Holland with the yachts and bring the Queen’s wine to the Nore” “Dunkirke Squadron” “Holland ord. to bring the Queen’s wine to the Nore” “Gone Convoy to the East Country” “At the Nore” “Sheerness Refitting” “Downes ordered to Scarboro to bring a ship to the Nore” (fitting for Newfound- 1 Navy List, January 1, 1707—-December 31, 1709, ADM 8/10. 31 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS Period February 1, 1711-February 28, 1711 March 1, 1711—March 31, 1711 April 1, 1711—April 30, 1711 May 1, 1711-May 31, 1711 June I, 171I-June 30, 1711 July 1, 1711-September 30, 1711 October 1, 1711—October 31, 1711 November 1, 1711-November 30, 1711 December 1, 1711-December 31, 1711 January 1, 1712-March 4, 1712 March 5, 1712-June 30, 1712 July 1, 1712-July 31, 1712 August I, 1712—-September 30, 1712 October 1, 1712-October 31, 1712 November 1, 1712-November 30, 1712 December 1, 1712—December 31, 1712 January I, 1713-January 27, 1713 February 1, 1713-February 28, 1713 March 1, 1713—March 31, 1713 April 1, 1713—May 31, 1713 June 1, 1713-June 30, 1713 July 1, 1713-July 31, 1713 August 1, 1713-September 30, 1713 October 1, 1713-October 31, 1713 November 1, 1713—December 31, 1713 VOL. 131 Duty “Spithead, ordered to the Downes” “In Holland to return to the Downes” “Tn Holland to bring the Queen’s wine to the Nore” “Sheerness, ordered to the Downes” “Ously Bay, for Russia” “Gone Convoy to Russia” “Coming from Russia. the Ist Convoy” “at the Nore” “Sheerness refitting” 2 (No record found) “Convoy between Folmouth and Spit- head” “Cruizing for the homeward bound Virginia ships” “Cruising off the Orcades for the homeward bound Virginia ships” “Nore, ordered to the Downes” “Downs, ordered to Bayonne with the Transports” “Gone to Bayonne with transports for soldiers” “Gone to Bayonne for some prisoners of war” “Coming from Kinsale to Plyo. to fit and repair to Spithead” “Kinsale, ordered to Plymo., to fit and repair to Spithead” “Plymouth refitting and ordered to Spithead” “Spithead, ord." to Guernsey and Jer- sey for some disbanded men of Mor- daunts Regiment.” “Gone to Gurnsey and Jersey for some disbanded soldiers” “Spithead cruizing between Start and the Isle of Wight” “At Portsmouth to fit and cruiz be- tween ye Start and the Isle of Wight” “To intercept the traders cruizing be- tween Start and the Isle of Wight” ® 2 Tbid., January 1, 1710—-December 31, 1711, ADM 8/11. 8 Tbid., March 5, 1712—December 31, 1713, ADM 8/12. Constant patrol of these waters was maintained in an effort to prevent the smuggling of wool from Britain to the Continent, the export of raw wool being absolutely forbidden at this time. NO. 2 Period January 1, 1714—March 31, 1714 April 1, 1714-April 30, 1714 May I, 1714-June 30, 1714 July 1, 1714-September 30, 1714 October 1, 1714—October 31, 1714 November 1, 1714—December 31, 1714 January I, 1715-January 31, 1715 February 1, 1715-March 14, 1717 April 1, 1717—October 31, 1717 November 1, 1717—December 31, 1717 January 1, 1718-March 25, 1718 March 26, 1718—March 31, 1719 April 1, 1719—April 30, 1719 May 1, 1719-December 31, 1720 January 1, 1721-April 30, 1721 May 1, 1721-January 31, 1722 February 1, 1722—-May 31, 1722 June 1, 1722-August 31, 1722 September 1, 1722-December 31, 1725 January 1, 1726—-January 10, 1728 January 11, 1728-March 31, 1728 April 1, 1728-April 30, 1728 May 1, 1728-July 31, 1728 August 1, 1728-December 31, 1728 January 1, 17290-January 31, 1729 February 1, 1729-April 1, 1729 April 2, 1729—-May 31, 1729 June 1, 1729-July 31, 1729 August 1, 1729-August 31, 1729 September 1, 1729-September 30, 1729 October 1, 1729-December 31, 1729 January 1, 1730-March 31, 1730 April 1, 1730-April 30, 1730 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 33 Duty “To intercept the traders cruizing be- tween the Start and the Isle of Wight” “At Portsmouth fitting and then re- turns to her station” “In the Downes going to Port Mahon with a storeship.” “Gone to Port Mahon with a store- ship” “Coming from Port Mahon” “At Shearness refitting” 4 “ordered to be laid up and paid off” (Not in service) “Baltick Squadron” “Hosp. Ship Looe” “To be paid off and laid up . . . Dep- ford” (Laid up) “Hosp. Ship Looe” “Mediterranean Squadron” “Looe .. 125 men 30 guns” (refitted as warship?) “Mediterranean Squad- ron” “Mediterranean Squadron” “Port Mahon” 5 “In the Mediterranean” “Ordered home from Mediterranean” (Lists missing ) (Not in lists, laid up) & (Laid up) “At Woolwich” “At Longreach” “Nore” “Downes to examine ships” 7 “Woolwich, not sheathed” (Laid up) “Woolwich fitting for Channel Serv- ice” “Downes” “Downes—Channel” “At Diep ordered to Spithead” “At Spithead” “Portsmouth Harbour” “Portsmouth Harbour, refitting for Channel service” 4Ibid., January 1, 1714—December 1, 1714, ADM 8/13. 5 [bid., January 1, 1715-April 30, 1721, ADM 8/14. 6 Tbid., May 1, 1721—-December 31, 1725, ADM 8/15. 7 Ibid., January 1, 1726-December 31, 1728, ADM 8/16. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS Period May 1, 1730—May 31, 1730 June 1, 1730—June 30, 1730 July 1, 1730-July 31, 1730 August 1, 1730-October 31, 1730 November 1, 1730-January 31, 1731 February 1, 1731-May 31, 1731 June 1, 1731-July 31, 1731 August 1, 1731-August 31, 1731 September 1, 1731-November 30, 1731 December 1, 1731—December 31, 1731 January I, 1732—January 31, 1732 February 1, 1732-February 28, 1732 March 1, 1732-March 31, 1732 April 1, 1732—-April 30, 1732 May I, 1732-May 31, 1732 June 1, 1732-December 31, 1734 January 1, 1735-May 4, 1735 May 5, 1735-June 30, 1735 July 1, 1735-July 31, 1735 August 1, 1735-August 31, 1735 September 1, 1735-October 31, 1735 November 1, 1735—April 30, 1737 May 1, 1737—May 31, 1737 June 1, 1737—December 31, 1739 January 1, 1740-December 31, 1741 January 1, 1742—January 3, 1742 January 4, 1742-April 30, 1742 May I, 1742-May 31, 1742 June I, 1742-June 30, 1742 VOL. 131 Duty “Spithead—Channel” “Gone with the money and clerks to Plymouth” “Ordered to fitt for Gibraltar, Spit- head” “Gone to Gibraltar as convoy to the transports” “To remain in the Mediterranean” “Cruizing on the Coast of Barbary” “Cruizing on the Coast of Sallee” “Ordered Home” “From the Coast of Sallee” “Portsmouth Service” “Stationed between the Start and the Isle of Wight” “Stationed between the Start and the Isle of Wight” “Ordered to the Downes” & “Start and Isle of Wight” “Portsmouth, ordered to Plymouth” “At Spithead, Channel Service” “Gone to Plymouth with money and clerks” “Nore” (Not listed, laid up) ® (Not listed, laid up) “Sheerness, fitting for the Channel” (Again as a hospital ship) “At the Nore” “At Spithead” “Gone to Lisbon with Sir John Norris” “At Lisbon” “To be paid off” (Not listed, laid up) 1° (Not listed, laid up) 14 refitting for Channel (Laid up) “Loo, 44 guns... Earl of Northesk Commander, Depford fitting for Channel Service” “At the Nore to convoy transports from the Downes to Spithead” “At Portsmouth refitting for channel service” 8 Tbid., January 1, 1729-December 31, 1731, ADM 8/17. 9 Ibid., January 1, 1732—December 31, 1734, ADM 8/18. 10 Tbhid., January 1, 1735-December 31, 1739, ADM 8/109 and 8/20. 11 Tbid., January 1, 1740—-December 31, 1741, ADM 8/21. NO. 2 July 1, 1742-August 31, 1742 September 1, 1742-September 30, 1742 October 1, 1742-October 31, 1742 November 1, 1742—November 30, 1742 December 1, 1742-December 31, 1742 January 1, 1743-March 31, 1743 April 1, 1743-April 30, 1743 May 1, 1743-May 31, 1743 June 1, 1743-June 30, 1743 July 1, 1743-July 31, 1743 August I, 1743-August 31, 1743 September 1, 1743-April 30, 1744 May 1, 1744- LAST CRUISE OF H.M.S. ‘““LOO”—PETERSON 35 “Cruizing 50 leagues off Capte Finis- terre” “Plymouth refitting for Channel Sery- ice” “In St. George’s Channel for 3 weeks” “Plymouth” “Spithead to clean at Portsmouth” “Portsmouth to cruize from 30 to 50 Igs. W.S.W. of Cape Clear for 6 weeks” “Plymouth, refitting for channel sery- ice” “To cruize between Bibao and St. Jean de Luz, to intercept some Caracca ships expected at St. Sebastian. . . to cruize 6 weeks on the Station” “Cruizing between Bilbao and St. Jean de Luz to intercept some Caracca ships expected at St. Sebastian . . .” “Portsmouth refitting for North America” “To attend on So. Carolina” “South Carolina” (No entry.12, The Loo had been lost February 5, 1744. Word apparently did not reach the Admiralty clerk keeping the navy lists until sometime in April. Entries of the location of ships were made on the first day of the month. ) 12 Tbid., January 1, 1742-May 1, 1744, ADM 8/22, 8/23, and 8/24. APPENDIX B LETTER FROM CAPTAIN ASHBY UTTING TO THE ADMIRALTY REPORTING THE LOSS OF THE “LOO” Port Royall 15th February, 1744. I am extremely sorry this should be the messenger of such disagreeable news as the loss of H.M.S. Loo. Will you please acquaint their Lordships that on the 4th day of February I was cruising on the station 8 leagues from the Cape of Florida when about 8 in the morning I saw a sail which I gave chase to and about noon spoke with her, she being an English “Snow”?! from Havannah and Missippy, but sailed by Frenchmen and two Spaniards, one that had been lately taken from the English and carried into Havannah. I having two men on board which was taken in her and the master having no copy of the condemnation and nothing to show for the sale but a common receipt. I seized her for the proprietors and was designed to send her into Charlestown but at the same time an Irish gentleman, a mer- chant that I had sent for on board, heaved a large packet overboard, which my boat took up and when opened found it full of French and Spanish papers, I then determined to see her in myself and also took her in tow. By the time I made sail it was 6 in the evening at which time the Pan of Mattances 2 bore S b E, the wind being SE. I steered NE b N till 12 at night by which time I was well assured I was got to the northward of the double Head Shott,? then hauled up NE. Till this time I was on deck myself and when thought I was passed all danger went and sat down in the cabin (as Doe assure you I did not go to bed one night in six of the time I was cruising here). At a } past one in the morning, the officer of the watch sent down to let me know he was in the middle of brakers and must Doe him the justice to say he behaved like an exceedingly good officer for before I was got upon deck which could not be ten moments, he had put the helm a Lee and the ship was at stays,* just as we hauled the main top sail the ship struck abaft but she pay’d off so far as to haul the head sails,5 when the Captain ® came and told me the tiller 1A brig having a small trysail mast set astern of the mainmast. The trysail was a fore and aft sail with a gaff and, in some cases, a boom. 2 A high, flat-topped hill lying inland from Matansas Bay on the northern coast of Cuba, a point on which mariners take bearings in setting a course up the Florida Straits (see fig. 1). 8 A group of keys lying in the eastern end of the Salt Key Bank which ex- tends to the center of the southern end of the Florida Channel (New Bahama Channel). 4 A vessel is said to be “ at stays” when heading into the wind in tacking. 5 Swung off from the wind so far that the head sails were caught across the wind pushing the bow of the ship around toward the reef. 6 The sailing master. 36 NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 37 was broke short off the ship, continued striking, I ordered all the boats out as far as possible. Immediately after he came and told me the rudder was gone and that she made some water in the hold but not much, we set all the pumps to work as you must believe on this occasion, and rather gained on her. By this time we was getting the long boat out when there came three or four severe seas and bulged? her immediately and had 5 foot water in the hold; I ordered the master and gunner to come and save what bread and powder they could before the water was over all, which they did and saved 20 bags of bread and 6 barrels of powder ® which was all we could save. By this time the “Snow” which shared the same fate, was on her broad side, the ship striking much and tareing all to pieces, and having no prospect of getting her off, ordered the masts to be cut away and all the upper deck guns and anchors to be thrown overboard, that she might lay quiet and by that means save the men which by good fortune she did, though all this time thought I was got on the double head Shott Bank when at daylight to my great surprise we was getting on a small sandy key about 1./1/2 cables® length long and ./1/2 broad which lay on the edge of the Bank of the Martiers 3 leagues!® without them and lies from Cape Florida WSW 7 or 8 leagues is quite steep too, we hav- ing no ground at 50 fathom right up and down not 10 minutes before the ship was ashore and is the only dangerous place on the Florida shore and Doe assure you that from the day I got on that station, I always had the Drapsy 1! Line going every 1/2 hour from 6 at night till daylight in the morn, the only reason I can give for finding myself on the Florida shore when I expected I was on the double Shott Bank which lies from each other SE b E and NN b N at least 16 leagues, must be occasioned by a very extraordinary and very uncommon new current; as soon as was daylight I landed all the men (but those that was em- ployed to scuttle the decks12 and get what water and what provisions we could, but could get but 2 butts out the whole day); at 10 o’clock this morning being Sunday we saw a small sloop when I immediately man’d and armed all the boats and sent them with orders to board her at all events and bring her here as she would be the means of carrying us off this dismal place, which I plainly saw that any common sea beat all over it and would certainly wash us all off, it being so low and dare not venture upon the main for the Indians which on this part of Florida are savages and innumerable, the next morning being Mon- day the boats to our great joy brought the sloop to us, the Spaniards having _all deserted her, she being about 25 or 30 tuns (at most) this day was employed in getting what provisions and water we could out of the ship with what men I could get to work which was but a few, though it was for all their good but all frightened and wanted to be gone for fear of the Indians and was very rebellious and mutinous dividing into parties and growling amongst themselves 7 “Bilged”—stove in her planks at or below the waterline of the ship. 8 A ship of 44 guns on foreign service normally carried 163 barrels of gun- powder in 1781 (Montaine, Will, The Practical Sea Gunners Companion, p. 73, London, 1781). ® The cable was 200 yards or one-tenth of a nautical mile. 10 The English and American marine league is equal to 3 nautical miles. 11 Utting was speaking of the dipsey line, which is the deep-sea lead line. 12 To cut openings in the decks. 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, I3I1 that they was all on a footing then; and they was as good as anybody and that everything was free. I thought it was best not to take any notice but prepared for our going as soon as possible. Here I found the service of the Marines whereof which was under command and did their duty extremely well and Centenells being obliged to mount 25 Marines and 25 seamen every night, though the place so small the Indians having numbers of canoes. Tuesday was employed getting what water we could out of the ship and put- ting the boats and sloop in order. Raised the long boat a strack 13 which enabled her to carry 60 men. On Wednesday being the 8th about noon I embarked all the men (which with the “Snow” included, was 274) viz 60 in the long boat, 20 in the barge,!4 ro in the yaul!® and 184 in the little sloop not 30 tuns; sent the sloop, long boat and yaul to 3 or 4 miles from the shore. After they were gone I went on board the wreck with what barrels of powder we had saved (except a little we took in each boat) and laid in proper places on the Star- board side the gun deck, that side being out the water by her heelding off to Port and proper Fewell!® in every place we could when all being laid, about 2 in the afternoon, I set her on fire and rowed off to the boats we kept in sight of her till sunset and she having blown up in several places and was in flames fore and aft but am much afraid the guns and anchors will fall into the hands of the Spaniards as they have frequent correspondence and trade with the In- dians, and it is my opinion the people which left the sloop we took, were over to Cuba in a launch directly to give them intelligence. This is the fatal end of H.M.S. Loo exactly as it happened. I sent in the long boat Mr. Randall and Mr. Lloyd, my first and third Lieuts., the Master in the barge and a mate in the yaul, myself and 2 Sevts. being in the sloop which when the hold was as full of men as could possibly be stowed, the deck was the same and much in danger of oversetting if any wind. I ordered them to follow me and in case of separa- tion to make the best of their way over to the Bahama Bank for Providence. I carried a light for them all night but as they all outsailed me they kept so far ahead that I lost sight of them by 12 at night when I was obliged to tack and stand to the northward, which did after making the proper signal but at daylight could see nothing of them, and it blowing fresh and the sloop top heavy with men could not carry sail, so obliged to bear away and take my fate through the Gulph of Florida for any part of Carolina even for St. Augustine (if I could fetch nowhere else) rather than all be drownded which Doe assure you had very little other prospect but thank God met with exceeding good weather and arrived at Port Royall on the 13th February at night and don’t in the least doubt but the boats are got safe to Providence long before this. I had two of the best pilots on board for the Gulph of Florida in all America who insist on it there could not be a better course steered and I have been numbers of times through the Gulph and am better acquainted with that and the coast of Carolina than any part of the world and had I had the honour to have command 20 sail and steering through for all our lives, should have steered the same course or rather more northerly, which as I observed before I cannot 13 The addition of planks to the gunwhales of the boat increasing her capacity. 14 The commanding officers boat. 15 A heavy double-ended row boat. 16 Fuel, combustibles. NO. 2 LAST CRUISE OF H.M.S. “LOO’’—PETERSON 39 account for but by some uncommon and very extraordinary current; as fast as I can get my officers together shall send home their depositions. Some of them seem inclinable to go to the West Indies and some to the Merchant Service and some home. I have sent home the Mate of the Watch with the Log Book and my Second Lieut. was on board the Snow when cast away but I should have been very happy to have found Captain Hardy 17 not sailed that I might have come home directly but as I can’t be now and there being a great many chances against my coming home in a merchant ship, without being carried to Spain,18 has deter- mined me to stay here till some opportunity offers to come home in a Man-of- War or if I could possibly be indulged with a court martial in America, I shall think it the greatest favour and if found by the Court, which I hope I shall, that I have done my duty as an officer on this unforseen unhappy affair, to me as well as to his Majesty’s Service, beg their Lordships will be pleased to give me leave to rely on their goodness for my being employed again on His Majesty’s service. I have enclosed the deposition of Mr. Wm. Lyford, one of my pilots who has sailed the Gulph of Florida for many years and beg their Lordships will be pleased to let somebody enquire of General Oglethorpe for his corretor.19 I am your most humble servant, Ashby Utting. Port Royal 15th February 1743/4. 17 Of the Rye, which ship had been ordered to England as convoy to merchant ships going from the Carolina colonies. 18 The risk of capture of unprotected merchant ships was very great. 19 Recommendation of Lyford. APPENDIX C MEMBERS OF THE CREW OF THE “LOO” ON HER LAST CRUISE The following is a list of names of all the men and officers appearing on the paybook of the Loo on her last cruise. Only those marked with an asterisk were on the ship when she was wrecked. In those days ships’ crews and officers were paid only at the end of a commission period, or at the end of a cruise. Men or officers who were transferred during a cruise were given a ticket by the purser showing the pay that was due them. These tickets were supposed to be held until the payday of the ship was announced in the newspapers, when they were presented at the designated pay office and the men received their pay. In actual practice, the interval of time between the issuance of the ticket and the actual payday was so great that many sailors suffered actual want, and to obtain funds, sold their tickets at enormous discounts to speculators. *Adam, Ervin *Adeane, W. Aiken, James Allen, Thomas Allman, John Anderson, James Angelo, Rogero *Arthur, James *Atkinson, Thomas Bagster, John *Baker, William Balderson, William Ball, William Balls, Benjamin Banke, John Baptista, John Barnes, William Barsey, Richard Barsey, Thomas *Bartlett, Joseph *Basham, Charles Bates, John 40 Quartermaster Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Landsman Quarters Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Master Able Bodied Seaman Able Bodied Seaman and Quartermaster’s Mate Surgeon’s Mate Able Bodied Seaman Quarter Gunner Able Bodied Seaman Able Bodied Seaman Landsman 2d Master’s Mate Able Bodied Seaman Gunner’s Mate and Quartermaster Able Bodied Seaman Deserted September 2, 1742 Deserted October 14, 1743, in South Carolina Deserted November 10, 1743, at Port Royal, S. C. Deserted March 30, 1743, at Plymouth, England NO. 2 Bates, William *Beckworth, Francis Belitha, Warren Bennet, John Bennett, Thomas Benson, Moses *Bent, John *Berry, John *Berry, Rowland Best, W. *Biggs, Thomas *Billonge, Jacob Birch, Robert *Bird, Richard *Bishop, Robert *Black, John Blackburn, John Blancher, Noah Bogue, Henry *Bond, Henry Boswell, David *Bould, William *Bousher, Walter Bradshaw, John Breamer, James *Briggs, William Briskingham, William *Bristoll, George Broughton, F. *Brown, James *Brown, Joseph *Brown, Nathaniel *Brown, Talbert *Brown, William *Buckley, John *Bugless, Ralph *Bugless, Stephen Bull, John Bullman, William Burdock, John Burns, Patrick Able Bodied Seaman Able Bodied Seaman Ist Lieutenant Able Bodied Seaman Landsman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Master’s Mate Boatswain’s Servant Able Bodied Seaman Carpenter Able Bodied Master Able Bodied Able Bodied Able Bodied Able Bodied Seaman Able Bodied Seaman and Coxswain Able Bodied Seaman Seaman Seaman Seaman Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Servant Able Bodied Master’s Mate Able Bodied Seaman Able Bodied Seaman and Quarter Gunner Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Captain’s Servant Sailing Master’s Serv- ant Able Bodied Seaman Captain’s Servant Able Bodied Seaman Seaman Seaman LAST CRUISE OF H.M.S. “LOO”—PETERSON 41 Deserted November 27, 1743, at Port, Royal;: SziC: Deserted September 2, 1742 Deserted December 109, 1742, at Portsmouth, England (Returned) Deserted April 13, 1743 Deserted April 6, 1743, at Plymouth, England Deserted September 16, 1742, at Plymouth, England Deserted October 15, 1743, in South Carolina Deserted April 13, 1743 42 SMITHSONIAN MISCELLANEOUS COLLECTIONS Burrough, John Burroughs, John Burt, William Burthen, James *Burton, John *Bushnell, William Butchard, Samuel *Butcher, Richard Campbell, Allen Campbell, Edward *Campbell, John *Canton, William Carrol, Michael *Carroll, John Carter, James Carter, Samuel Cartwright, Benjamin Carty, John *Caunter, Henry *Chandler, Edward *Charming, Edward Charter, William Chippendall, Jona *Christopher, William *Churton, James Collins, John Compton, John *Conday, Richard Condray, Charles Conner, John Cook, John Cook, John Cook, John Cook, Thomas *Cormick, James *Cormick, John Cormick, Michael Cormick, William *Couch, James Courteney, F. Able Bodied Seaman Able Bodied Seaman Midshipman Able Bodied Seaman Able Bodied Seaman Carpenter’s Mate Able Bodied Seaman Able Bodied Seaman Captain’s Servant Corporal Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Captain’s Servant Able Bodied Seaman Steward and Ordinary Seaman Able Bodied Seaman Boatswain’s Mate Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Landsman and Able Bodied Seaman Quarter Gunner Captain’s Servant Able Bodied Seaman and Midshipman Ordinary Seaman Able Bodied Seaman Gunner’s Servant Ordinary Seaman Able Bodied Seaman Midshipman Able Bodied Seaman Surgeon’s Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman and Ordinary Sea- man Clerk Deserted October 5, 1743 at New York Deserted October 4, New York Deserted October 13, 1743, in South Carolina Deserted September 16, 1742, at Plymouth, England Deserted April 13, 1743 Deserted November 17, 1743, at Port Royal, S. C. Deserted December 14, 1742 VOL. 131 NO. 2 Coverley, William Cowe, Peter *Cowen, Philip Cowen, William *Cowey, Robert Cox, Anthony Creese, John *Crilly, Thomas Cross, Samuel *Crow, Philip *Crowley, Bryan Cunnan, John *Curry, John Davidson, Alexander Davies, Griffith Davies, Matthew *Davies, Thomas Davies, William Dawson, William Day, Joseph Deacon, J. *Dean, James Delancy, Lawrence *Demount, J. Dent, Digby *Dickson, William Dixon, David Donnaly, Sam *Donnovan, John Douglas, David -Douglass, Robert Dove, Benjamin Dover, Saunders Dowes, William Downing, Robert Downing, Thomas *Dowsing, Samuel Driscoll, John *Driscoll, William Drisdall, Alexander Duncan, John Ordinary Seaman Ist Lieutenant Able Bodied Seaman Able Bodied Seaman Midshipman Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Captain’s Servant Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Captain’s Servant Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Captain’s Servant Cook Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Captain Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Carpenter’s Servant Corporal Captain’s Servant Captain’s Servant Trumpeter Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman and Master’s Mate Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Carpenter’s Mate LAST CRUISE OF H.M.S. “LOO’’—PETERSON 43 Deserted September 2, 1742 Deserted October 13, 1743, in South Carolina Deserted August II, 1743, at Plymouth, England Deserted October 28, 1742, at Plymouth, England Deserted October 4, 1743, at New York Deserted December 12, 1742, at Portsmouth, England Deserted March 30, 1743, Plymouth, England Deserted April 13, 1743 Deserted December 12, 1742, at Portsmouth, England Deserted April 6, 1743, at Plymouth, England Deserted September 10, 1742 44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 Able Bodied Seaman Able Bodied Seaman Surgeon’s Mate Able Bodied Seaman *Duncan, William Dunn, George *Dunn, Matthew Dunstar, James Deserted October 5, 1743, at New York Able Bodied Seaman Able Bodied Seaman Landsman Dyar, Anthony *Dyer, Darby Eades, Thomas Deserted September 2, 1742, at Plymouth, England Able Bodied Seaman Able Bodied Seaman Master’s Servant Landsman *Easton, Thomas *Edgecombe, William Ellis, Benjamin *Ellory, Robert Ervin, Adam Evans, George Fanson, Andrew *Farmer, William Able Bodied Seaman Captain’s Servant 3d Lieutenant and 2d Lieutenant Able Bodied Seaman Midshipman and Mas- ter’s Mate Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Farrel, James Ferguson, Robert Field, John *Fisher, William Fitzgerald, Morris Fitzpatrick, Henry Deserted December 12, 1742, at Portsmouth, England Fletcher, John Ordinary Seaman *F ling, Timothy Able Bodied Seaman Forrest, Richard Able Bodied Seaman Forsith, William Able Bodied Seaman *Forster, Randal Able Bodied Seaman Fortiene, Joseph Able Bodied Seaman Deserted November 17, 1743, at Port Royal, S. C. Fraser, Daniel Able Bodied Seaman Deserted November 27, 1743, at Port Royal, S. C. Frost, George Sailmaker and Mid- shipman *Fullmore, Henry Boatswain Gally, Thomas Gibson, George Gilbert, Thomas *Gilmore, Arthur Gilmore, John *Gold, William Goldsmith, John *Good, John Master’s Servant Able Bodied Seaman Able Bodied Seaman Quarter Gunner Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Surgeon’s 2d Mate Deserted April 16, 1743, at Plymouth, England Deserted April 13, 1743 Goodsides, Abraham Gordon, James Gordon, Robert Midshipman Deserted October 8, 1742 Midshipman NO. 2 Gorman, David Graham, Matthew *Graham, Samuel Grant, Henry *Green, Henry Gregory, Jo. Gresham, Charles Grossier, John Grun, Nicholas *Hallet, John Hamilton, James Hampshire, William Hancock, George Hancock, John *Harman, William Harris, John Harris, Richard Harris, Thomas Harrison, Theodore Hartie, John *Hartman, Christian *Harwood, Lewis *Hatch, John *Hatfield, Willey Hawkins, John Hawkins, John Hay, David Hayes, Samuel Hays, James Headley, Christian Heaver, James Hemins, John *Henderson, James Henry, William Henton, John *Hickey, Thomas *Higgenson, James *Higginson, John *Highmas, Thomas *Hinds, Michael *Hogg, Peter Holliday, Richard Holmes, John *Hope, John Horlock, Joseph Able Bodied Seaman Coxswain Able Bodied Seaman 3d Lieutenant’s Serv- ant Ordinary Seaman Master’s Servant Landsman Landsman Able Bodied Seaman Second Gunner Midshipman Able Bodied Seaman Able Bodied Seaman 3d Lieutenant’s Serv- ant Able Bodied Seaman Boatswain’s Servant Pilot Master’s Servant Able Bodied Seaman Ordinary Seaman Able Bodied Seaman 2d Gunner and Master at arms Able Bodied Seaman Captain’s Servant Ordinary Seaman Master’s Mate Able Bodied Seaman Able Bodied Seaman Trumpeter (?) Able Bodied Seaman Midshipman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Boatswain’s Servant Quartermaster’s Mate Master’s Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Midshipman Quartermaster Surgeon’s 2d Mate LAST CRUISE OF H.M.S. “LOO’—PETERSON Deserted September 2, 1742 Deserted September 10, 1742 Deserted June 30, 1743, at Portsmouth England Deserted October 13, 1743 Deserted December 26, 1742 Deserted April 13, 1743 46 SMITHSONIAN Horseman, Arthur Hughes, John *Hull, Lancaster Hulsal, Arthur Hunt, John Hunt, John Hussey, I. *Hyslop, Thomas Ingerton, Dennis Jackman, William Johnson, Luke Jolly, George *Jolly, Thomas Jones, Anthony Jones, David Joynes, James *Juba, Luke Julian, Charles Keeler, Robert *Keighley, William *Kelley, Morgan Kelsey, William *Kennedy, John Kent, Ambrose Keys, Robert Killrick, Isham *Kilsey, William King, Daniel *King, Nathaniel *Kingsbury, William *Kirk, James *Kirk, Samuel *Kivey, John *Knowles, Edward *Knowling, James Lake, Mark Lamb, William Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman and Midshipman Able Bodied Seaman Sailing Master Captain’s Servant Armorer Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Ordinary Seaman 3d Lieutenant’s Serv- ant Able Bodied Seaman Able Bodied Seaman Quarter Gunner Purser’s Servant Able Bodied Seaman Master at Arms Able Bodied Seaman Able Bodied Seaman Surgeon’s Servant and Able Bodied Sea- man Able Bodied Seaman Quartermaster Able Bodied Seaman and Midshipman Captain’s Servant Landsman Able Bodied Seaman Captain’s Servant Gunner’s Servant Gunner Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ist Lieutenant’s Sery- ant Quarter Gunner MISCELLANEOUS COLLECTIONS VOL. 131 Died March 24, 1742, at Plymouth, England Deserted December 12, 1742, at Portsmouth, England Deserted August II, 1743, at Plymouth, England Deserted April 13, 1743 Deserted April 6, 1743, at Plymouth, England Deserted December 19, 1742, at Portsmouth, England Deserted November 17, 1743, at Port Royal, S. C. Deserted September 2, 1742, at Plymouth, England NO. 2 *Lander, Robert *Langston, Richard *Lather, Samuel Lavermore, Joseph Lawler, John *Ledgerwood, James *Lemarr, Stephen *Leslie, George Lewis, Christopher *Lewis, Edward Lewis, Theodore *Lewiswentz, David *Limb, Giles Linch, William Liston, William *Lloyd, William Lobb, Thomas Lodge, John Logan, Charles Long, Samuel Long, William Lookert, Patrick Luch, John *Lyford, William *Maby, John Malt, Issac Manley, George Mannon, John *Marriot, William Martin, John Maxwell, James May, William *McCann, John McCarty, Daniel McChownley, Lott McCleland, Robert *McClockland, William McCowley, Charles McDugal, James McKensie, Samuel McKnight, James *McNeal, William Meachem, James Able Bodied Seaman Quartermaster Able Bodied Able Bodied Able Bodied Able Bodied Able Bodied Purser Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Captain’s Servant Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Midshipman Surgeon’s Servant Boatswain’s Servant Ordinary Seaman Seaman Seaman Seaman Seaman Seaman Able Bodied Seaman Bodied Bodied Seaman Seaman Able Bodied Seaman Able Bodied Seaman and Midshipman Ordinary Seaman Able Bodied Seaman Able Able Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Quartermaster Ist Lieutenant’s Serv- ant and Able Bodied Seaman Able Bodied Able Bodied Able Bodied Able Bodied Seaman Seaman Seaman Seaman Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Quarter Gunner Ordinary Seaman Quartermaster Able Bodied Seaman LAST CRUISE OF H.M.S. ‘“‘LOO’—PETERSON Deserted April 16, 1743 Deserted April 13, 1743 Deserted November 5, 1742, at Plymouth England Deserted March 30, 1743, at Plymouth, England Deserted October 2, 1742 Deserted October 3, 1743, at New York Deserted April 13, 1743 Deserted October 4, 1743, at New York 48 SMITHSONIAN MISCELLANEOUS COLLECTIONS Mellan, John Meridith, John Mignam, George *Miller, James *Miller, Nicholas Miller, Stephen Mills, Nathan *Millsom, James Mitchell, R. *Molineaux, John Moore, Samuel More, Thomas Morran, William Morrison, John Morrow, Hugh *Mortimer, Robert Murphy, Francis *Murphy, James *Murphy, John Murphy, Patrick Murphy, Samuel *Nagan, James *Nellson, Robert *Newgent, Patrick *Newson, Richard Nobbs, Stephen Norman, Edward *Norman, William Northesk, Earl of *Nott, Ebinezer *Nuikle, Robert Ogburne, John Oliver, Richard *Orr, Archibald *Osbourne, William Parker, Peter Parr, William *Parrott, Samuel Patrick, Richard Patten, Hugh Pearse, Samuel Boatswain’s Mate Ordinary Seaman Surgeon Able Bodied Seaman Able Bodied Seaman Boatswain’s Servant Able Bodied Seaman Ordinary Seaman Ordinary Seaman and Able Bodied Sea- man Yeoman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Captain’s Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Landsman Ordinary Seaman Gunner’s Servant Captain Ordinary Seaman and Able Bodied Sea- man Able Bodied Seaman Captain’s Servant Landsman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Quartermaster Captain’s Servant Cook’s Servant Able Bodied Seaman Able Bodied Seaman Deserted April 13, 1743 Deserted September 2, 1742 Deserted September 10, 1742, at Plymouth, England Deserted August II, 1743, at Plymouth, England Deserted December 29, 1743, at Port Royal, S. C. Deserted September 2, 1742 Deserted December 12, 1742, at Portsmouth, England Deserted March 30, 1743, at Plymouth, England Deserted April 16, 1743 VOL. 131 NO. 2 *Pearson, Thomas Pegan, Roger Pegan, Thomas Pelican, John Pemell, Thomas *Peters, William Phallem, Edmund *Phillips, Thomas *Phonix, Philip *Pickering, Michael *Plantain, Charles Poole, Thomas Potterfield, George Powell, Philip Pownswell, Edward *Pretty, Thomas *Price, James *Priest, Lewis Puttick, William *Quaco, John *Quin, John Ramsey, Patrick *Randall, James *Randall, James *Randall, John Randell, James *Randolph, J. *Ratsey, George Reed, Thomas *Rhode, John *Richards, Nicholas *Richardson, William Richey, David *Richmond, Andrew *Rider, Charles Ordinary Seaman and Able Bodied Sea- man Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman and Midshipman Able Bodied Seaman Cook Able Bodied Seaman Clerk Able Bodied Seaman Pilot Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Captain’s Servant and Ordinary Seaman Able Bodied Seaman 2d Gunner 2d Lieutenant and Ist Lieutenant Able Bodied Seaman Able Bodied Seaman and Midshipman Ist Lieutenant Able Bodied Seaman and Master’s Mate Cooper and Able Bod- ied Seaman Able Bodied Seaman Able Bodied Seaman and Ordinary Sea- man Able Bodied Seaman Able Bodied Seaman Quartermaster’s Mate Able Bodied Seaman Able Bodied Seaman LAST CRUISE OF H.M.S. “LOO’—PETERSON Deserted November 27, 1743, at, Port Royal, S/C: Deserted March 17, 1742, at Plymouth, England Deserted March 30, 1743, at Plymouth, England Deserted December 12, 1742, at Portsmouth, England Deserted October 6, 1743, at New York 49 50 SMITHSONIAN MISCELLANEOUS COLLECTIONS Risden, Joseph Roach, Henry Roberts, David Roberts, Edward Roberts, John Roberts, John Roberts, Joseph *Roberts, William Robinson, Alexander Robinson, John Rogers, Robert Rowe, Edward Rowe, George Rowe, Peter Royall, James *Russell, Richard Ryan, John St. Lawrence, Samuel Salisbury, Edward *Salmon, Joseph *Salter, James *Saunders, Joseph Scannel, John Scott, George Scott, Richard *Shaw, Patrick *Shearing (John or Joseph) *Shearing, Thomas Sherwood, William Shirley, Washington *Shoart, Oliver Simmonds, Richard *Simms, James Singleton, William *Skinner, Philip Sluman, John Smith, Benjamin *Smith, James Smith, Paul Smith, Theodore Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Sailmaker’s Mate Captain’s Servant Captain’s Servant Captain’s Servant Able Bodied Seaman Landsman Able Bodied Seaman Quarter Gunner Able Bodied Seaman Able Bodied Seaman Boatswain’s Servant Able Bodied Seaman Ordinary Seaman Able Bodied Seaman Ist Captain’s Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Gunner Landsman Able Bodied Seaman Ist Lieutenant’s Servy- ant and 3d Lieuten- ant’s Servant Able Bodied Seaman Able Bodied Seaman 3d Lieutenant Able Bodied Seaman Quarter Gunner Ordinary Seaman Ordinary Seaman Ordinary Seaman Gunner Able Bodied Seaman Able Bodied Seaman Servant Boatswain’s and Able Seaman Servant Bodied VOL. 131 Deserted December 12, 1742, at Portsmouth, England Deserted October 28, 1742, at Plymouth, England Deserted December 19, 1742, at Portsmouth, England Deserted September 2, 1742 Deserted December 12, 1742, at Plymouth, England Deserted October 4, 1743, at New York Deserted September 10, 1742, at Plymouth, England Deserted April 13, 1743 Deserted April 6, 1743, at Plymouth, England Deserted September 2, 1742 Deserted October 4, 1743, at New York NO. 2 Softley, Robert Somerwel, Joseph Southard, Thomas Spare, Thomas Sparks, Joseph *Spinks, Stephen Spry, Nicholas Stanford, Richard Stapleton, A. Stephenson, James *Steuart, Neil Stevenson, Henry Stewart, Alexander *Stiveash, Stiven *Stoneham, John Stradder, Forbel Stroud, John *Sullivan, Thomas Swain, William *Sweeny, Daniel *Swickman, Thomas Swift, Theodore *Tabler, Thomas *Taylor, John *Taylor, Jonathan *Taylor, William Thatcher, John *Thompson, Richard Thoyer, Peter *Tipper, John Tobyn, George Todd, Alexander *Torginton, Joseph Treacey, William Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Quartermaster’s Mate Landsman Surgeon Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Landsman Boatswain Ordinary Seaman 3d Lieutenant’s Serv- ant and Captain’s Servant Able Bodied Seaman Landsman Able Bodied Seaman Captain’s Servant and Able Bodied Sea- man Carpenter’s Servant and Able Bodied Seaman Able Bodied Seaman Captain’s Servant and Ordinary Seaman Quartermaster Pilot Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Carpenter’s Servant Able Bodied Seaman LAST CRUISE OF H.M.S. “LOO’’—PETERSON Deserted March 30, 1743, at Plymouth, England Deserted April 13, 1743 Deserted September 10, 1742, at Plymouth, England Deserted March 17, 1742, at Plymouth, England Deserted September 2, 1742, at Plymouth, England Deserted October 15, 1743, in South Carolina Deserted April 13, 1743 Deserted August 5, 1743, at Portsmouth, England Deserted October 6, 1743, at New York Destered March 30, 1743, at Plymouth, England Deserted December 26, 1742, at Portsmouth, England 52 SMITHSONIAN MISCELLANEOUS COLLECTIONS *Triming, Thomas *Trist, Nathaniel Trott, Thomas Trotter, Benjamin Trouve, Paul *Trunker, William Tunis, Michael Turford, Thomas *Utting, Ashby Vincent, Aaron *Vivian, John *Wadlin, Richard *Walker, James Walker, John Walker, Peter Wallis, Benjamin Warceys, Thomas *Ward, Samuel *Weatherill, John Wedlock, John Welch, Michael Welsh, Peter *Wemuss, Robert Wemy, James Wheeler, Robert *White, Richard *White, Robert Whitver, Thomas Wilkinson, Edward *Williams, Richard *W illmot, Theodore Wills, J. Wilson, George Wilson, Lawrence Wise, John Woodgate, William *Wotton, William Wright, William *Yeates, James *Young, John Younger, John Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman 3d Lieutenant’s Sery- ant Able Bodied Seaman Able Bodied Seaman Carpenter Captain Able Bodied Seaman Carpenter Ordinary Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ordinary Seaman Cook’s Servant Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Able Bodied Seaman Ist Lieutenant’s Sery- ant Able Bodied Seaman Able Bodied Seaman and Midshipman Able Bodied Seaman Able Bodied Seaman Captain’s Servant Ordinary Seaman Captain’s Servant Deserted September 10, 1742, at Plymouth, England Deserted October 2, 1742, at Cork, Ireland Deserted December 12, 1742, at Portsmouth, England Deserte@ September 16, 1742, at Portsmouth, England Deserted December 12, 1742, at Portsmouth, England Deserted September 23, 1742, at Plymouth, England VOL. 131 MARINE LIST Colonel Cotterel’s Regiment : *Allen, Robert Baker, John *Ball, William *Bond, George *Brooke, Joshua *Brooks, George Brooks, John Brown, George Brown, John Cant, Thomas Clint, Richard *Cole, Joseph *Cook, Samuel Corbett, Thomas Diamond, Robert *Dight, Edward - *Douglass, Edmund, Sergeant Drake, Richard *Farries, William, Sergeant Finch, Samuel *Fitzsimmons, Thomas *Gaddish, Lazarus *Gleddon, Richard *Gould, William Grovenor, Francis Hall, William *Hardeman, William *Harding, Matthew *Haydon, Timothy Hodge, Hugh *Hold, Thomas ~ Hughes, Edward *Hyatt, Samuel Isaac, Titus Johnson, Adam *Jones, Philip *King, Joshua Lee, John Martin, Thomas *Matthews, Joseph *McCraw, Daniel Miller, Richard Mitchell, John LAST CRUISE OF H.M.S. “LOO’’—PETERSON 53 Deserted December 26, 1742, at Portsmouth, England Deserted June 30, 1743, at Portsmouth, England Deserted December 12, 1742, at Portsmouth, England Deserted December 2, 1742, at Portsmouth, England 54 SMITHSONIAN MISCELLANEOUS COLLECTIONS *Morgan, John *Morgan, William *Mould, Thomas *Murray, Peter *Overshott, John Pearce, George Phillips, Edward Phillips, William Prest, Timothy *Ridghewothh, Thomas *Risden, Samuel Roberts, John Rowls, John Searle, John Short, George Spraeg, Nicholas *Stevens, Samuel *Stone, John Thrasher, John *Toll, John Trovana, William *Turner, Jonas *Turpin, John *Vaughan, Hector, Lieutenant *Walker, William Whiteker, Thomas *Williams, Thomas Wills, Samuel *W oodeson, John Woodley, Nicholas Colonel Wynyard’s Regiment Aldridge, William Chappel, Samuel Clayton, Joseph Hancock, Justinian Hope, Richard *Kent, Benjamin *Korgett, John Parker, Daniel *Stokes, William *Trowes, Richard *Wright, Samuel VOL. 131 APPENDIX D ROSTER OF THE FAMILY AND SERVANTS OF GEORGE CLIN- TON, GOVERNOR OF NEW YORK FROM SEPTEMBER 1743 TO OCTOBER 1753, PASSENGERS ON THE “LOO” DURING HER LAST VOYAGE TO AMERICA, AUGUST-SEPTEMBER 17431 Clinton, Ann (Governor’s wife) Cunningham, Archibald Clinton, Ann Davies, Ann Clinton, George (Governor ) Ellis, Thomas Clinton, Henry Fenton, Sarah Clinton, Lucy Harvey, Phillip Clinton, Mary Ryves, James Ascough, John Vanham, John Aurneo, Ann White, Margaret Blundell, Christopher Williams, William Catherwood, John Willson, Ann _ SOON ANDwW DH 1 Based on the Muster Roll of the Loo, ADM 36, vol. 1823. APPENDIX E COMMANDING OFFICERS OF THE “LOO” MP arland MiNOUEEE: cae cbichasisioie cae cieines ore April 4, 1707—October 1, 1709 PETOR DEER A OMAN ojnis, oe ctesaiecte e w/tnavouse aoaierere October I, 1709- ? PRAVV AGUA ES EAtt: cya /scrorsiciaeielererais, a eranie chorale 1700- 1715 PREM IAME ori Miad wacats's slo wcee 3. ar oierd 5 vera esd eainisie April 1717(?)—December 1717 POP EOEHEHE, GEGERE Kies diaisiae wed onercvess April 1718- 1720 SCOte—(C CAntAA! oalay.c nae ove stnsteg sci October 1720- 1722 Waterhouse, Tho. (Commodore)....... February 1727/28—-October 1728 Me Mather laces J cle dmis tee cate e October 22, 1728-April 2, 1729 PBetmeley, VV MISARD ws in coicnles von e baa es April 2, 1729- ? BUESCRE GNOME 2rd Bers Bietercls aeileln'ecsieainm ne May 5, 1735-May 1737 pe Obeasic, BAG) Ol eisio ic ca acted aveie's cieccle es January 4, 1741/42-September 1742 MPSPCRIE MOPESISY 8 tc ghee caioes hiaiaionorvera-a arate hoch September 17, 1742-April 11, 1743 oN ECMO CEM OMLIND igs eh nO ccls! prceistioieinl veh eleyecw des April 11, 1743-February 5, 1743/44 1 Based on the Navy List, 1707-1744. 55 aed a, | ' J ae a er, rl y DMS ie ye -= ee a SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 2131, NO: 2, PL. 2 1. Copper coins recovered from the wreck site of the Loo in the summer of 1950. Left to right: Spanish 8 maravedi circa 1600, Spanish 8 maravedi circa 1640, and Swedish 4 ore dated 1720. This last coin was important evidence in dating the wreck. Lent to the U. S. National Museum by Dr. and Mrs. George Crile, Jr. 2. Small Queen Anne pewter teapot from the wreck of the Loo. Giit to the National Museum from E. A. Link. The pot was damaged by fire when the Loo was burned by her captain after being wrecked. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 2315, NO. 2; PES The remains of the ship lay between two coral reefs in (Photograph by Dr. George Crile, Jr.) he wreck site. 35 feet of water. The salvage boats of the expedition at anchor over t 2, Pees NO. id, ONIAN MISCELLANEOUS COLLECTIONS S SMITH oY} STOAOI (‘Af ‘apg 95 loot) “Aq Aq Y Jeu} UOTPESHAIUS pues-[e1oo ou} OJON deis ‘eu ojo ) ‘yyeys Joyoue dy} Jo pus sy} pue sul 9 94} WO} surpnsajosd s1oyoue JaMOd S,007 IY} FO BUO jo SULI OU] ‘say Aq ydeisojoy ) ‘sounqord UOTJOU JOJEMIIPUN SUTYR} St “Yfoy oy} ye “4 { ‘QHD 9841094) “ACT “sport 0} 1IoJeBM JO Jol e UVM WW0}}Oq purs 9} YSeM ST 1d}UdI ou} ul JOAIP 9U,L (Afi faptay as100r) ) Jews JOAOOUN ‘OUIS YOIIM OY} UO SUTIYIOM SIDATC] SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131-"NO.. 2, PL. 6 A basket of coral-sand-encrusted shot coming up from the bottom. (Photograph by Dr. George Crile, Jr.) 2, Pla 131, NO. VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS *MOLIE PROIG 94} JIM poxreur ‘pauval>s puke poysnsoua-pues-[e109 Yoys punod-zr :woyoq ‘uns JOAIMS Jopunod-% ve 10} JoYS IO Aopunod-g eB 1Oy adeIs o[suIs Be “Joys punod-F “‘yystr Suns JOATMS Jopunod-1 ve Jo} JOYS 10 Jopunod-z1 wv 10; odevis o[suts B Joys punod-1 Yo :4o}UVD “MOAIe PROIq 9Y} 9}JOU + pouRalI pue Po SNAIUI-PueS-|eIOI ‘yous punod-g : doy, ‘00'T IY} FO 901M out jo OPIS ou WUOT] podtIAOIIT JOUS UOAT-Jse-) 27 Fea NO. VOL. 131, SMITHSONIAN MISCELLANEOUS COLLECTIONS yo dumny e peppequa ‘urepooi0d [BIOS 0} poyseze “1oy}OUY ‘Z -aqqj0q Watds sseys UddIS-Y1ep B JO uorjsod wio}j0g ‘g ‘[e109 Jo dun] e ul ulejaotod asoury) Jo JusWIseIyy *S ‘oT WOq pads ssejs use18-yAep & JO YN ‘Ph “voHRIOIIp any{q asouIy) JO SpuUsUUSeIy “E *9]}0q aids sse[S Usais-y1ep B Jo uosod woyog ‘Z ‘UOT}eIOSIp oN] ‘QIVMIUOJS IZETS-}]LS YSISUG JO SJUOUISeIF OMT “I + OPIS YIM JY} WOT} PoJOAOIIT S]PLIayeUL ITWUIBII-) VOL. 132, \NOi>2;; (RES) SMITHSONIAN MISCELLANEOUS COLLECTIONS ‘oTpuey yod ssvig ‘g ‘aUO}S 94} JO pud Jo] oy} UO opETq v Aq jJo] SYIVUT dJoU :aUO}s SuTUadIeYsg “4 “oUog Stq “€ “drys IY} JO $210}S Jwatu-}Rs dy} WIZ soUog MOD ‘9 pur ‘S ‘FY ‘Z ‘T :9}IS YIM oy} WAY patoAOIo1 SyD9!qGQO 10 PE. JL. 137, NO. V¢ SMITHSONIAN MISCELLANEOUS COLLECTIONS “dIys oy} JO Satoqs }eotUI-}]eS (‘af ‘aptig a8t00r ‘sapy Aq ydessojoyd ) JY} JO SUIeIII oy} Wort MeL ssid & Jo UOTAOd B SuIpjoy puey sowel NO. 2, PL.-ag VOL., 131, SMITHSONIAN MISCELLANEOUS COLLECTIONS “SULSSII OY} WOT; YOoY uost-jYysSnorM ‘Z ‘feo. yo dun] ‘g “yoo0] Joop uoray ‘S ‘adid asmey Jo taddnos & Ayqeqo.id ‘adid pray ‘h “yyoq [[NY uosAt-yYSNoIA\ ‘EC “Suissa Surpurys s.diys ay} wo.sy yfoq -oAGf *Z “UMOUYUN asn JON ‘9dVIG UOAI-JYSNOAM ‘1 9S YIM IY} WoOI, pataaodas sypolqO NO; 2; (PL. 12 VOL. 131, SMITHSONIAN MISCELLANEOUS COLLECTIONS ‘aired *“MOIIeE PeOtd ot [HY eB WOT} pee] } 3 ol I S JYSI ‘ dy} YYM poyseur ‘qn} Jo yseo e wor dooy sse4q : XBpeOsq po snsoua-pu es-[e1oo ‘FVYye] + OUTS ‘ de Dal ‘ yIIIM Ul0}JO dy} WOT pataaoovat syoofqO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131, NO. 2, PL. 2g Ee. A. Link rigging the main boom of the Blue Heron to raise a 1-ton iron barrel from the wreck site. (Photograph by Mrs. George Crile, Jr.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. LL, Rar luc Heron at Marathon, ) the Loo being unloaded from the 6 aph by Dr. George Crile, Jr.) A coral-sand-encrusted iron 6-pounder from ( Photogr Crile, Jr.) e Oo Ss (Photograph by Mrs. Geor Coral-sand crust falling away from an iron 6-pounder. 16 od NO. Lil; VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS (‘Af ‘optaD 965100: ) ‘sayy Aq ydeisojoyd) ‘“poAowod sem Suvod purs-]e109 Jaye ‘0O7T ay} Wosy Japun’ MI-Q SUOT /*. ee af rae goene: ~ - Ss SMITHSONIAN MISCELLANEOUS COLLECTIONS ‘dooys ystureds poanjdeo e pavoqe “Fri ‘1 ArenNAqaJ doy PAATIIe 0O7T IY} JO AIIM 9} JO SIOAIAINS OOL pue SUI) “de ‘ PAD Cire ; $9 14 a" oe ~~ a& oilee i 1 61@ee ag vel ® js. stnc ys wont ae AGOMTAMTR aT cent “unial)ong ngage Sa inhale + boo tebe een tes a) MU oo Pea eee’ 5715.5 P*, anteroposterior diameter: transverse diameter.............20e-ees 6,057.3 M’, anteroposterior diameter: transverse diameter across anterior por- CIE ccs ooctoee tae oo sera «getts vie cia.sias ook eiteielnie Rant ae aerate 6.2: 8.5 M?, anteroposterior diameter: transverse diameter across anterior por- tonnes Pr rere hed rue” ay ee cbr at Be rene Ree 6.2: 9.1 M®, anteroposterior diameter: greatest transverse diameter............. 4.9: 7.5 Length of lower cheek tooth series P: (at alveolus)-Ms, inclusive...... 428 Length of lower premolar series, P: (at alveolus)-Ps, inclusive......... 23.3 Length of lower molar series, M:i-Ms, inclusive............eeeseeeeees 20.0 Ps, anteroposterior diameter: transverse diameter.............-e+esee-- 5-5: 3.2 P,, anteroposterior diameter: transverse diameter..................4-- 6.7: 4.1 M,, anteroposterior diameter: transverse diameter of talonid........... 6.5 sigan M;, anteroposterior diameter: transverse diameter of trigonid.......... 6.5: 5.4 Ms, anteroposterior diameter: transverse diameter of trigonid.......... 6.7: 4.5 NO. 7 MAMMALIA FROM THE ALMY FORMATION—GAZIN 13 ECTOCION, cf. OSBORNIANUM (Cope), 1882 A right lower jaw fragment with P,-M, (U.S.N.M. No. 20645) has teeth more robust than in the jaw belonging to the E. ralstonensis skull, and in addition Ps is more progressive with a better developed talonid basin and a large entoconid. P,, moreover, has a rather dis- tinctly developed paraconid or parastylid. This specimen is tentatively regarded as representing the Gray Bull species E. osbornianum. There are in addition three other jaw fragments, each with a comparatively large molar which may likewise be referred. Although Granger (1915) recognized three species of Ectocion in the Clark Fork beds, including both E. ralstonensis and E. osborn- tanum, Simpson (1937b), in his treatment of the material, believed (except for rare E. parvus) that a single species was represented in which there was a shift in the mean size, the length of M, for ex- ample, between successive horizons from Clark Fork to Lost Cabin time. While this seems evident in the demonstration given, I am, nevertheless, concerned about the more progressive P, in the larger Almy specimen. The character of P, might likewise show marked variation within a species, but there are three small-toothed or E. ralstonensis specimens which have P, preserved, and in each of these this tooth is distinctly less progressive. The correlation may be a coincidence, but if not, I am inclined to believe that in this instance a distinct species is actually represented. MEASUREMENTS IN MILLIMETERS OF TEETH IN THE SPECIMEN OF Ectocion, cf. osbornianum, U.S.N.M. NO. 20645 P,, anteroposterior diameter: transverse diameter of talonid........... PASis Cay M,, anteroposterior diameter : transverse diameter of trigonid.......... 6.8: 5.9 M2, anteroposterior diameter: transverse diameter of trigonid.......... mT O.5 PHENACODUS ALMIENSIS Gazin, 1942 Plate 2, figures 3 and 4 A relatively small species of Phenacodus, P. almiensis, is clearly the most abundantly represented form in the fauna. The 32 speci- mens in the collection referred to it comprise about 44 percent of the total. The type specimen, U.S.N.M. No. 16691, consists of maxillae with the canines and P* to M* in a scarcely worn state, together with certain limb and vertebral portions. Although collected in 1941, it remains after six subsequent collecting trips the best specimen of this species extant. I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I P. almiensis is much smaller than Phenacodus p. intermedius but appreciably larger than P. copei. It differs essentially from the Phena- codus primaevus group, other than in size, in better developed ex- ternal styles, particularly the parastyle, and in exhibiting slightly more crescentic cusps. The protocone in the upper molars, for example, is united by better defined crests to the protoconule and metaconule and generally with the hypocone as well. The metaconule is about on a line between the metacone and hypocone, not posterior to this, as frequently observed in P. primaevus, nor so forward as in Ectocion. P. almiensis is significantly larger than any of the P. copei material observed, and although the latter exhibits fairly prominent external styles on the upper molars, the cusps, particularly the protocone, have less developed crests than in P. almiensis. Granger (1915) noted that the metaconules were weak or absent in P. copei. These are appar- ently not reduced in P. almiensis. Moreover, the upper premolars, strangely enough, appear more advanced than in P. copet. P® has a well-defined and separate tritocone, described as weak in P. copet, and this tooth in P. almiensis also has incipient to clearly defined conules and tetartocone. P* is distinctly molariform in appearance, and is recognized among isolated teeth by the absence of a mesostyle and by the somewhat less developed, though by no means weak, hypocone (or tetartocone, in upper premolar nomenclature). Both conules are present and well defined. Compared to earlier species, P. almiensis is distinctly larger than P. matthewi, as well as P. gidleyi, and not nearly so robust as P. grangeri among the species known from the Colorado Tiffany. Moreover, the teeth are relatively not so broad transversely as in P. grangeri. The premolars are decidedly more advanced than in Phenacodus bisonensts. As noted earlier (Gazin, 1942), the teeth of P. almiensis show some resemblance to Ectocion in the development of the external styles and somewhat crescentic appearance of the cusps; however, I do not believe that Ectocion is represented because of the markedly elongate (anteroposteriorly) and relatively narrow upper molars, the position of the metaconule, and the comparatively unreduced condition of the hypocone of M%. Also, in the lower molars the anterior crest joins both the protoconid and metaconid, and the hypoconulid is not so close to the entoconid as it usually is in Ectocion. NO. 7 MAMMALIA FROM THE ALMY FORMATION—GAZIN 15 MEASUREMENTS OF UPPER TEETH IN SPECIMENS OF Phenacodus almiensis U.S.N.M. No. 16691 U.S.N.M. Type No. 21286 Length of cheek tooth series, P*-M*, inclusive.......... 42.7% 41.0% Length of molar series, M*-M®, inclusive............... 26.1% 26.7 P*, anteroposterior diameter: transverse diameter....... 8.2: 7.4 77"'!S.6 P*, anteroposterior diameter: transverse diameter....... 8.5: 8.5 7.9: 9.2 M’, anteroposterior diameter: transverse diameter *..... 9.0: 10.0 9.0: II.0°% M*, anteroposterior diameter : transverse diameter *..... 922113 8.8 12'5* M®, anteroposterior diameter: transverse diameter *..... 7.7: 10.8 8.93 1035 @ Approximate. * Anteroposterior diameter of upper molars taken perpendicular to anterior margin and transverse diameter across anterior portion. PHENACODUS PRIMAEVUS Cope, 1873 Plate 2, figure 5 Two specimens in the collection may well represent typical Phena- codus primaevus. One of these, U.S.N.M. No. 21287, is a lower jaw with P; to Mag, inclusive, and the other an incomplete lower molar. The teeth in No. 21287 are comparable in size to those in the Clark Fork material referred to P. primaevus. The length of the lower molars is near the lower limit of the range given for each (Simpson, 1937b, p. 18) and the widths are nearer the upper limit, suggesting relatively broad teeth, not otherwise distinguished from P. primaevus. About eight specimens of smaller size, though not comparable to P. almiensis, correspond in general proportions to Gray Bull materials earlier regarded as Phenacodus intermedius. The dimensions of teeth in one of these (U.S.N.M. No. 20644), evidently the largest of the group, are given in the accompanying table. In this and others having comparable lower molars the teeth are observed to be relatively slen- der, particularly in comparison with the larger, broad-toothed form discussed above. A single specimen encountered by Simpson (1937b, p. 19) in the Clark Fork collections, representing a smaller group which approximates the intermediate-sized form in the Almy fauna, was regarded by him as Phenacodus primaevus, small var., cf. inter- medius. The Almy materials may be treated in a similar manner, for taxonomic convenience, because, although the limited Almy materials might appear to be clearly defined, I find it difficult to distinguish P. intermedius from P. primaevus in the Gray Bull collections. Never- theless, I feel rather strongly opposed to a concept which recognizes more than one subspecies of the same form coexisting in time and at the same geographic locality. 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I MEASUREMENTS IN MILLIMETERS OF CERTAIN LOWER TEETH IN JAWS OF Phenacodus Pp Ck. Px D3 P. almiensis intermedius primaevus U.S.N.M. U.S.N.M. No. 20643 No. 20644 No. 21287 P,, anteroposterior diameter ......... 9.6 TTS 12.7 Ps, transverse diameter of talonid..... 6.2 7.8 10.2 Mz, anteroposterior diameter......... Q.1 11.8 Eur M,, transverse diameter of talonid.... 7.4 9.8 in? M2, anteroposterior diameter ......... seers 3 theia 12:55 M2, transverse diameter of trigonid.... ... ana 12.6 @ Approximate. DINOCERATA UINTATHERIIDAE PROBATHYOPSIS?, sp. Plate 1, figures 2 and 3 Two upper premolars, possibly both P*, or P* and P*, but of differ- ent individuals as indicated by wear, are evidently of Probathyopsis. They are, however, significantly larger than Probathyopsis praecursor Simpson (1929) of the Clark Fork beds. They correspond closely in size to a P® belonging with a partial skeleton of cf. Bathyopsis fissi- dens Cope from the New Fork upper Wasatchian (Gazin, 1952, p. 64), but are slightly more brachydont. The unworn Almy premolar (U.S.N.M. No. 21283) measures 14.6 mm. long perpendicular to anterior margin by 16.5 mm. wide perpendicular to outer wall. P* in the type of P. praecursor measures 11.5 by 13.3 mm. in the same directions. The proportions of the upper premolars are comparable to those of the earlier Bathyopsoides harrisorum Patterson (1939) from the Plateau Valley beds, although the unworn Almy premolar is a little shorter anteroposteriorly and broader transversely than the B. harris- orum P*. Nevertheless, the transverse lophs have about the same proportions. Although Patterson has indicated certain differences in cusp pattern of M2, it would seem from the evidence presented by Dorr (1952, p. 89) that Bathyopsoides is possibly a male Probath- yopsis. Better evidence with regard to this situation should be forth- coming in the more detailed study of the Hoback Basin material contemplated by Dorr. NO. 7 MAMMALIA FROM THE ALMY FORMATION—GAZIN 17 REFERENCES Core, Epwarp D. 1873. Fourth notice of extinct Vertebrata from the Bridger and Green River Tertiaries. Palaeont. Bull. No. 17, pp. 1-4. 1882. Contribution to the history of the Vertebrata of the lower Eocene of Wyoming and New Mexico, made during 1881. 1. The fauna of the Wasatch beds of the basin of the Big Horn River. Proc. Amer. Philos. Soc., vol. 20, pp. 139-191, I fig. Dorr, JoHN A. 1952. Early Cenozoic stratigraphy and vertebrate paleontology of the Hoback Basin, Wyoming. Bull. Geol. Soc. Amer., vol. 63, pp. 59-94, figs. 1-6, pls. I-7. Gazin, C. LEwIs. 1942. Fossil Mammalia from the Almy formation in western Wyoming. Journ. Washington Acad. Sci., vol. 32, No. 7, pp. 217-220. 1952. The lower Eocene Knight formation of western Wyoming and its mammalian faunas. Smithsonian Misc. Coll., vol. 117, No. 18, pp. 1-82, figs. 1-6, pls. I-II. 1956. Paleocene mammalian faunas of the Bison Basin in south-central Wyoming. Smithsonian Misc. Coll., vol. 131, No. 6, pp. 1-57, figs. 1-2, pls. 1-16. The occurrence of Paleocene mammalian remains in the Fossil Basin of southwestern Wyoming. Journ. Paleont. (in press). GRANGER, WALTER. 1915. A revision of the lower Eocene Wasatch and Wind River faunas. Part 3—Order Condylarthra. Families Phenacodontidae and Menis- cotheriidae. Bull. Amer. Mus. Nat. Hist., vol. 34, art. 10, pp. 320- 361, figs. 1-18. JePsEN, GLENN L. 1930. Stratigraphy and paleontology of the Paleocene of northeastern Park County, Wyoming. Proc. Amer. Philos. Soc., vol. 69, pp. 463-528, figs. 1-4, pls. I-10. MattTHeEw, WILLIAM D. 1915a. A revision of the lower Eocene Wasatch and Wind River faunas. Part 1—Order Ferae (Carnivora), Suborder Creodonta. Bull. Amer. Mus. Nat. Hist., vol. 34, art. 1, pp. 4-103, figs. 1-87. 1915b. A revision of the lower Eocene Wasatch and Wind River faunas. Part 4.—Entelonychia, Primates, Insectivora (part). Bull. Amer. Mus. Nat. Hist., vol. 34, art. 14, pp. 429-483, figs. 1-52, pl. 15. PATTERSON, BRYAN. 1939. New Pantodonta and Dinocerata from the upper Paleocene of west- ern Colorado. Geol. Ser. Field Mus. Nat. Hist., vol. 6, No. 24, pp. 351-384, figs. 100-111. ScHuLtz, ALFRED R. 1914. Geology and geography of a portion of Lincoln County, Wyoming. U. S. Geol. Surv., Bull. 543, pp. 1-141, figs. 1-8, pls. 1-11. Srmpson, Georce G. 1928. A new mammalian fauna from the Fort Union of southern Montana. Amer. Mus. Nov., No. 297, pp. 1-15, figs. 1-14. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I 1929. A new Paleocene uintathere and molar evolution in the Amblypoda. Amer. Mus. Nov., No. 387, pp. 1-9, figs. 1-9. 1933. Braincasts of Phenacodus, Notostylops, and Rhyphodon. Amer. Mus. Nov., No. 622, pp. 1-19, figs. I-3. 1935. The Tiffany fauna, upper Paleocene. 3.—Primates, Carnivora, Con- dylarthra, and Amblypoda. Amer. Mus. Nov., No. 817, pp. 1-28, figs. I-14. 1936. A new fauna from the Fort Union of Montana. Amer. Mus. Nov., No. 873, pp. 1-27, figs. 1-16. 1937a. The Fort Union of the Crazy Mountain field, Montana, and its mammalian faunas. U. S. Nat. Mus. Bull. 169, pp. 1-287, figs. 1-80, pls. I-10. 1937b. Notes on the Clark Fork, upper Paleocene fauna. Amer. Mus. Nov., No. 954, pp. 1-24, figs. 1-6. EXPLANATION OF PLATES PLATE I PRIMATES, ANACODON ps AND PROBATHYOPSIS? FROM THE ALMY PALEOCENE Fig. 1. Anacodon? nexus, new species: M: and M2 (U.S.N.M. No. 21282), type specimen, occlusal view. Natural size. Figs. 2 and 3. Probathyopsis?, sp.: 2, Upper premolar (U.S.N.M. No. 21283), occlusal view; 3, upper premolar (U.S.N.M. No 21284), occlusal view. Natural size. Fig. 4. Carpolestes, cf. dubius Jepsen: Ps (U.S.N.M. No. 21280), labial (left) and lingual views. Six times natural size. Figs. 5-8. Plesiadapis cookei Jepsen: 5 and 7, Left ramus of mandible (U.S.N.M. No. 20785), (5) occlusal view, twice natural size, and (7) lat- eral view, natural size; 6 and 8, left ramus of mandible (U.S.N.M. No. 16698), (6) occlusal view, twice natural size, and (8) lateral view, natural size. Fig. 9. Plesiadapis? pearcei, new species: Right ramus of mandible (U.S.N.M. No. 20787), type specimen, occlusal and lingual views. Three times natural size. Fig. 10. Plesiadapis rubeyi Gazin: Right ramus of mandible (U.S.N.M. No. 16696), type specimen, occlusal and lingual views. Three times natural size. PLATE 2 CONDYLARTHS FROM THE ALMY PALEOCENE Figs. 1 and 2. Ectocion ralstonensis Granger: 1, Skull (U.S.N.M., No. 20736), lateral and ventral views. Natural size; 2, left ramus of mandible (U.S.N.M. No. 20736), lateral and occlusal views (M: restored from right side). Natural size. Figs. 3 and 4. Phenacodus almiensis Gazin: 3, Right upper cheek tooth series (U.S.N.M. No. 16691), type specimens, occlusal view (M”* restored from left side) ; 4, left ramus of mandible (U.S.N.M. No. 20643), occlusal view. Natural size. Fig. 5. Phenacodus primaevus Cope: Left ramus of mandible (U.S.N.M. No. 21287), occlusal view. Natural size. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131,"NO. 7, PL. 1 PRIMATES, ANACODON?, AND PROBATHYOPSIS? FROM THE ALMY PALEOCENE (SEE EXPLANATION AT END OF TEXT.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.131,0NO:.7,, PES.2 CONDYLARTHS FROM THE ALMY PALEOCENE (SEE EXPLANATION AT END OF TEXT.) » a ‘Von ig mien Nae . a «A Prat me a i | | Lael p ; hip SL : ae eee "% ; pb, pre SA itsinin” ee, Ot ta WO Hy 1 Ui" aly 4 thee me, wat Pan's MP Who 1 sini era 7 . itp. er dial etl S Geet, id oy A ih i vir Veh § A eC tal yoy ae ¥ Ly (pr iy: | at petea a i PMNS t' Ti hea +e a . Real Che Mabe eeat') EB PV xe - Neel ey nN Ohh a Pee at ' BA 2. THE MAMIE AE NII OO 10g ie: IABWATHA : ais , | eh ; ol i or ie Teva . é eile oi, Ve , / a i om tae Bali / a | a | os a Ah PVT YE. ieAT a Gee), aslibc, oe Hee nee ‘Pipaivaniane ; Pa wiihy | SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 8 Charles D. and Mary Waux Galcott Research Fund THE GEOLOGY AND VERTEBRATE PLE ONTOLOGY OF WPPER "EOCENE STRATA IN THE NORTHEASTERN PART OF THE. WIND RIVER BASIN, WYOMING PART 2. THE MAMMALIAN FAUNA OF THE BADWATER AREA (WitH 3 PLaTEs) By C. LEWIS GAZIN Curator, Division of Vertebrate Paleontology United States National Museum Smithsonian Institution (PusiicaTIon 4257) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION OCTOBER 30, 1956 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U. S. A. CONTENTS Page MEIER MIE CAOIEE Vc xsl aha is sc 2 vor awie'e esis) sacs Om aeaitayalsiee steiaisie ale aitiers ¥ieie'¥ bles als el ai I NGI OW IEC SITIEMESA i cjeyatey st sjsheucts efe\ciovels oyoteraeroriavsteleinis eve ioisierslellciaverets) > s/arsle: syelsve 2 ELISEOGY OL MAMVESES ALI OM stovoa cts Sraleio sists le eis tavalelaalcfalcneisiclovele oie cieisievese' elevate clsiste 2 Occurrence and: preservation Of material . os cjccecescecne csc sicsocs vice 3 se) Frac Water Palsiai- tic.c.ctac ae.cts hod Sui aiealbi ale ah’e gist eisisie @ie's'e a's Siu/ae/e.0isieisais 4 Worselation and age Of the: fauna (5.502516 sos tebe sing. dees ceils nee oeisieaee 5 Systematic description of the Mammalia occ 6isc 60 anes s,ciecdiecsviewsessse 0 BLS AP MeANA A 8 dans crssnysy Spoke de tal eters opevetatmtarebe ofa ors! ale, <16, 0, diana sscim cieieie.e.daia 9 PVIGCIPINGAEL a5) soi, oiesisersae so aisle as,0 a)e sieie- 9.2.6 ¥-¥'6ia. 6 66ers sareeiae 9 APOE | osc aissc Bale weir ee OLR ewe we alelec der wiatkl ew ese.ave eo sisidieversierety 9 PROROTIAE.. (3.2 ctisscsrsiaim esc ae ersieeraais ous cbieisia's estein eye casa eens 9 CMSINE TAs Sc 03 ateua s & aj tia cehec tea ici tater here eie eieiaiait Oe whale ara whee ies ona ais weenie 10 ROPERNEV ON Pyaar ato ta' cts reha)watcasin at ue Syst faa, Dies valnl a stalere St elarasa a Goa abs) RO eoharsaxora'e 10 PAMINOCVOMMGAC a. ,sa ee ec cee ae Cots renin eae ey SONa Oa chap wers 10 MACTDAC siacs cikgtate ee a eiaine Macatee tle Oe eb WS ois Oks Rae STAG ales 10 MG Gotatiar cat ana 5. Siac < oie ao erevane atencreh rete Meare ee wee bial Sie'e od eahain ged bNlars II EA VOMSOUMGHUIGAC sic je-chavei ad Sra iencioha erence evailoreiaw ciate ola os e-cinnies aie acs II Pea ESSOUACES La nee) craiah seca ta bracaiil alavers re oie e eiata) Ba latadioiaiere gral antrene s-@:ersinl¥ javarovace 12 EET ae yet tela reel atetolatelord che chee toucloue elas areiene esa eveuerevelwiatwie mis Sisiae eke 12 EPEOMILOLACTIMLAG’, a.6:3)orstcin vie raters. oa 'ale eis vere a\e.a-b/ela piaibie(aldle wieleie/stereieiae 13 RTC EMP ENGARM, lrg i.' Sera.) tts Sevduare, dbrelerd ea ejhiale, daoeioain'e.o-ofsloune’s 13 VACUA Gee oto hus ceNaiere aaah e aisraleiwial o sina ais,auesateuayalesstPiasa a, Sie ersislale tains 16 EISWACOMONITICIAE vole tities wine tecveie tare vere ale aye GING, waracarOle Sade ees SSO 23 RE MMAAL EM s eteeheac wet ae static a Zea cis nik Ce Casale slone aaa ow aN Gia veces 23 DT CHODUMIG ACN iia Ne cree her iste SIS HE a ee ane erate fed 23 PN UC IOCRACLAL OF sts aieisto ie ashen teterals Sait oeiciaratoiola Soave wid oe e Miers aia aio 26 ORCOMEENIAAE WE. 12) «ors oroalassarapalataters sree chelate wu ernie ala Ricfaltrave ate ales atte 30 BENE ORIE TY CIM AC!seskue poeoetele AoyevasSecie ah oti er Ro nto ees phecoee ale eiesios Chloe 30 aR RIES POR Mars icici se oon, aspirant data, i srelie 6 Mydaneo gid'ns wv ayaudin ie ack meomaceeha nl eehzie 31 RMU POLAEMEMGTA LAE) SPARES 6 oyu chee cle ato nl eik sig crater Seore nie a sccit die oheiy a wi aitmid wradin's 34 ILLUSTRATIONS PLATES (All plates follow page 35.) ~ . Lagomorph, carnivore, condylarth, and perissodactyls from the Badwater upper Eocene. 2. Perissodactyls from the Dry Creek and Badwater upper Eocene. 3. Artiodactyls from the Badwater upper Eocene. CHART _ . Suggested phylogenetic arrangement of North American tapiroids..... 16 ‘ = FadReReutecests ss ert oes see-r b> - ~etehet alt ony chgathestinn, bon eben enn ot « am er ua TKO ti i y } os DORAN WOR. Owbald oh i aol ed ti ws NADER DETER ee po ee eee ees EMMETT ARADO TELOMERES) FLO a Oe w a bee he eo) s Sd @ Cad W eNOS a Pd el'Fr aes de spaces dyads th eieiny Olea Rewee iW See hava «oS Oe Gree Pweeee reese) 1 OD vabiiae yr Te Pee * a CoeN Pas) GATTO EG § CIM awig see eee © eh PR Oa bw wile ae © eae & ' v Sa Gh 1 eh es Gea G se ee aee ee i Foe caida cin Ge CY i at's tans TOCHPI OP PVOGiSMte his ed a ee) GPFHePaAPevicebewraes i ope bee ode rw ane Paul eanAMas SiEG eset 80h abe e anes a i | i pera aap =" Piwale mees-t Uy ma hent s > bm vimtimn yd ieee Trea: awhipeilé ,reeo™® (tes eoecas Vle- he & - cre ec: eee ' woe it Gite) ‘base 1a 10 4%. ts a e é i & we @ Tr’ a Diss 6s e¢s F@EC OR CE Ee PhDs 5d VL OTEV OE 4B? (epee Ses 9 Malet A ’ % ‘ * s* »«@ ** 4 & e4ab db eeheres®sod oan ede ee oe) n a ae ep Mies owe glee Oak) allie oiiemnnedy ® ae § % ede ts op@ee ne ee ee sm Pen eae reeset (Meee Jabaa ve ne . a aw + onng cote 5g eek Sialay Te « RMOMTAR Tedd fins ane ‘thal aacliag A svn -aepiee, wn ea hal tent aluan pe alle ee ¥ i ee slats - | : rT ; Charles D. and Mary Waux Walcott Research Fund THE «GEOLOGY ANDm«VERTEBRATE,: PA- PEONTOEOGY’T OR UEPER (LOCENE STRATA IN THE NORTHEASTERN PART OF THE WIND RIVER BASIN, WYOMING PART 2.1 THE MAMMALIAN FAUNA OF THE BADWATER AREA By C. LEWIS GAZIN Curator, Division of Vertebrate Paleontology United States National Museum Smithsonian Institution (WITH 3 PLATEs) INTRODUCTION The significance of the Wind River Basin in contributing informa- tion on mammalian faunas of upper Eocene time has been appreciated only during comparatively recent years. Although a rather meager fauna had been known from beds of Uintan equivalence below the Beaver Divide along the south side of the basin for many years, it is rather surprising that the occurrences on the north side were not earlier discovered, particularly in view of the long history of collecting associated with the adjacent lower Eocene Wind River formation. Discovery of the occurrence of upper Eocene mammalian remains along Badwater Creek near the site of the old Badwater Post Office by Wood, Seton, and Hares in 1936 was followed by investigations of others, notably those of Harry A. Tourtelot for the U. S. Geological Survey and parties for the Smithsonian Institution. The present study stems largely from an interest in Eocene tapi- roids, the upper Eocene representatives of which are so well repre- sented here, and is in part a sequel to an earlier review of artiodactyls 1 Part 1 of this paper is a study of the geologic relations, in preparation by Harry A. Tourtelot. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 131, NO. 8 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I of this age. Misunderstandings in the past as to the characteristics and, in some instances, the age of related types, largely from inadequate original descriptions of forms in these two ungulate groups, eariy confused the picture and resulted in incorrect identifications and evi- dently misleading conclusions as to the horizon represented by the Badwater assemblage. It is hoped that the present review and revision will clarify the record and render more useful the information to be derived from this occurrence. ACKNOWLEDGMENTS I am particularly indebted to Dr. G. Edward Lewis of the U. S. Geological Survey for relinquishing to me for restudy this very inter- esting assemblage, and assuring me that no embarrassment would ensue. I wish also to acknowledge the helpful information furnished me by Harry A. Tourtelot both in the office and in the field. Tourtelot, obligingly and with contagious enthusiasm, showed me the more sig- nificant and likely collecting sites during our 1946 field exploration. Investigation was immeasurably aided by my being permitted to examine and study related materials in universities and other museums, and by the loan of pertinent specimens in these collections for direct comparison. Acknowledgment is particularly due Dr. J. LeRoy Kay at the Carnegie Museum, Dr. Glenn L. Jepsen at Princeton University, Dr. George G. Simpson at the American Museum, and Dr. Joseph T. Gregory at Yale Peabody Museum. Dr. H. E. Wood, II, aided in furnishing information on materials in the original Badwater col- lection. The exquisite pencil drawings depicting selected materials shown in plates I-3 were prepared by Lawrence B. Isham, staff artist for the Department of Geology in the U. S. National Museum. Mr. Isham also prepared the chart showing the tapiroid sequence. HISTORY OF INVESTIGATION Discovery of upper Eocene vertebrate materials in the Badwater area was made by Wood, Seton, and Hares, as reported by them in 1936. Moreover, this would appear to be the first record for the oc- currence of upper Eocene on the north side of the Wind River Basin. Recognized by Wood, Seton, and Hares were Amynodon advenus, Telmatherium, cf. cultridens, and a crocodile. Collections later (1942) secured by J. D. Love and G. E. Lewis from Lysite Mountain to the north of Badwater Creek, for Yale University, include remains iden- tified by Lewis as Telmatherium, cf. cultridens, and an indeterminate no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 3 helaletid (this is Dilophodon). Nevertheless, significant collections, more representative of the fauna, were not obtained from these beds until Harry A. Tourtelot and his assistants secured for the U. S. Geological Survey in 1944 and 1945 the materials discussed in his maps and reports of 1946, 1948, and 1953. Identification of the Geological Survey material was made by G. E. Lewis and reported by him in 1947. Collections for the Smithsonian Institution were made by F. L. Pearce, Chester Gazin, and myself in 1946, and Pearce and I revisited the localities with good results in 1953. Other known col- lections include that made by A. E. Wood in 1948 for Amherst Col- lege, the small mammals represented having been described by him in 1949. Further collecting was done by Tourtelot and the unusual Malaquiferus tourteloti skull was found by him near Dry Creek in 1948. Materials secured by the U. S. Geological Survey also included a collection made by J. R. Hough in 1950, and in her 1955 report on the Sage Creek occurrence comparisons are made with portions of the Badwater fauna. OCCURRENCE AND PRESERVATION OF MATERIAL The principal occurrences for materials of the Badwater fauna are the low gray-green exposures along the south side of Badwater Creek between 24 and 34 miles almost due northwest of the site of the now abandoned Badwater Post Office. These are immediately to the south and to the southeast of the mouth of Clear Creek in the southeast part of section 14, the southwest part of section 13, and the northwest part of section 24, T. 39 N., R. 89 W. The above, together with other scattered localities, are shown on both the 1946 and 1953 maps of Tourtelot, as well as his map accompanying part 1 of this study. The discovery of Badwater vertebrate remains by Wood, Seton, and Hares was made at a locality south of Badwater P.O. about 3 miles to the southeast of the above exposures and in section 32, T. 39 N., R. 88 W. Determinable remains have likewise been encountered on Lysite Mountain to the north of the Badwater area by Lewis and Love, probably in section 25, T. 42 N., R. 90 W. The Dry Creek exposures, almost certainly the same age as those on Badwater Creek, are about 20 miles due west and include the sites for the Malaquiferus and Eomoropus skulls, in the NW4 sec. 14 and the SE sec. 9, re- spectively, T. 39 N., R. 92 W. Much of the fossil material encountered has been rather fragmen- tary although there are five comparatively good skulls in the National Museum-Geological Survey collections from there. Two of these are 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, I3I of Diplobunops and the others are of Protoreodon, Malaquiferus, and Eomoropus. It is particularly noteworthy that although often much fractured, the remains show almost no distortion, a condition unusual in collections of Eocene age, particularly those from the Uinta basin. The bone for the most part is a light buff color and the teeth amber to dark brown. The beds themselves do not resemble exposures of the Uinta formation in Utah but much more closely resemble the light gray-green middle Eocene reworked volcanic ash of the Bridger for- mation in southwest Wyoming. The upper Eocene fossil-bearing beds in the Badwater area have been named the Hendry Ranch member by Mr. Tourtelot and regarded as a part of the Tepee Trail forma- tion. Definition and description of these beds are included in part 1 of this paper. THE BADWATER FAUNA MARSUPIALIA?: Didelphidae ? : Peratherium?, sp. LAGOMORPHA : Leporidae : Mytonolagus wyomingensis A. E, Wood RODENTIA * : Paramyidae : Rapamys?, sp. Sciuravus dubius A. E. Wood Paramyid indet. (large) Paramyid indet. (small) Eomyidae: Protadjidaumo?, sp. Cricetidae : Cricetid indet. CARNIVORA : Limnocyonidae : Limnocyon?, sp. Miacidae: Miacis, cf. robustus (Peterson) CONDYLARTHA: Hyopsodontidae : Hyopsodus, cf. uintensis Osborn PERISSODACTYLA : Equidae: Epihippus, cf. gracilis (Marsh) Epthippus, cf. parvus Granger Brontotheriidae : Brontotheriid indet. Chalicotheriidae : Eomoropus anarsius, new species no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 5 Helaletidae: Desmatotheriwm woodi, new species Dilophodon, cf. leotanus (Peterson) Hyracodontidae : Epitriplopus?, sp. Amynodontidae* : Amynodon advenus (Marsh) ARTIODACTYLA : Dichobunidae : Pentacemylus?, sp. Apriculus praeteritus, new genus and species Agriochoeridae: Protoreodon, cf. petersoni Gazin Protoreodon, near P. pumilus (Marsh) Protoreodon pearcei, new species Diplobunops, cf. matthewi Peterson Oromerycidae: Malaquiferus tourteloti Gazin Leptomerycidae: Leptotragulus, cf. medius Peterson Leptoreodon?, sp. * Rodents are as described by A. E. Wood from material in the collections at Amherst College and not represented in collections at the U. S. National Museum. The Amynodon material is that identified by H. E. Wood, II, in the collection of Wood, Seton and Hares. CORRELATION AND AGE OF THE FAUNA The fauna listed above is, of course, by present standards upper Eocene in age. Moreover, there would seem to be no doubt but that it is Uintan. Apparently not any of the forms here recognized are characteristically or exclusively Duchesnean. The general association of forms seen in the assemblage and the development reached in certain groups such as the agriochoerids rather strongly suggest an upper Uintan stage close to that at Myton pocket. Considering first the reasons for not regarding the fauna as Duchesnean, only the rodent which A. E. Wood (1949) cited as questionably Protadjidaumo might be interpreted as this age. Never- theless, this form is also older than Lapoint in age, as Kay (1953, p. 24) cites it as occurring in the Randlett fauna, and the latter I re- gard as but scarcely distinct from that of Myton, including it ? (Gazin, 1955, chart 1) in the Uintan. The remainder of the fauna is composed of genera that so far as Duchesnean is concerned are characteristically earlier or common to both Uintan and Duchesnean. Thus, besides 2 As currently being proposed by the Committee on Nomenclature and Cor- relation of North American Continental Tertiary of the Society of Vertebrate Paleontology. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I Protadjidaumo, only Epihippus, Amynodon, and Protoreodon are common to the two, and the genera Mytonolagus, Rapamys, Eomoro- pus, Epitriplopus, Pentacemylus, Diplobunops, Leptotragulus, Lep- toreodon, and possibly Desmatotheriwm and Dilophodon are charac- teristic of the Uintan. Some of these almost certainly gave rise to later types in the Oligocene but the genera in the latter group are not ac- tually known in Duchesnean time. The genera Sciuravus, Limnocyon, and Hyopsodus are survivals from Bridgerian time and Miacis ranges through most of the Eocene. The evidence for regarding the fauna as upper rather than lower Uintan pertains to the presence of Mytonolagus, the possible Pro- tadjidaumo, and particularly to the stage of development shown in the Protoreodon and Diplobunops material. On the other hand, the pres- ence of Sciuravus (doubtfully this genus according to Wood), Lim- nocyon, and Eomoropus might suggest an earlier horizon, but these are comparatively rare forms in Uintan deposits and their absence heretofore in beds as, late as Uinta C is not nearly so significant as the fact that the agriochoerids are distinctly advanced over those of Uinta B time. Question as to whether the Badwater fauna should be correlated with that from Myton pocket or with that known from the Randlett member may well have little significance. Protadjidaumo is not known from Myton pocket but the Badwater specimens are stated by A. E. Wood to consist of incisors only, so can scarcely merit serious debate. Mytonolagus is known from both levels but the Badwater form is a different species. Dilophodon (“Heteraletes”’) might suggest a rela- tionship to the Randlett, but Uinta collections in the U. S. National Museum show that this form is present also in the Myton fauna. A slight evidence favoring the Myton fauna is seen in the artiodactyl species represented. Of the Badwater forms, Protoreodon pumilus is evidently present in all three occurrences, but P. petersoni and probably P. pearcei are known only in the Myton fauna. Also, the Diplobunops from Badwater resembles the Uinta form D. matthewi more closely than it does the broad-skulled D. crassus. It is entirely possible that, although a difference in stratigraphic level has been described for the Myton pocket and Randlett occurrences, the differ- ences that may be pointed out are of ecologic significance, as sug- gested by the rather different nature of the deposits. The beds at the Myton pocket and Randlett occurrences received sedimentary materials from quite different rock sources. I have been unable to detect any change which can be regarded as evolutionary between forms common to the two levels. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN a Lack of uniformity of opinion regarding the source of Douglass’s Eocene materials from the Sage Creek areas makes comparison with the fauna or faunas represented there decidedly unsatisfactory. I have not had the opportunity of studying the field occurrence firsthand so am unable to contribute any information to the stratigraphic picture. Nevertheless, from the materials that I have examined in the collec- tions of Kay and of Hough, understood to be from a single horizon in the Eocene sequence, I find a comparatively close relationship be- tween their fauna and the Badwater assemblage. While I do not con- cur in several of the identifications cited in Hough’s (1955) paper, nor do I agree with the Duchesnean age assignment, there would ap- pear to be a near equivalence in time, possibly also in environment, considering the similarity in faunal representation. With regard to the Douglass collection, I have seen only the helaletid and am reason- ably convinced that it represents an advanced dilophodont distinct from the Dilophodon in Kay’s collection. If, as Horace E. Wood (1934, p. 255) postulates, Douglass’s amynodont might have weath- ered from the overlying Cook Ranch Oligocene, it is not impossible that the dilophodont did likewise and is a distinctly small and perhaps unprogressive species of Protapirus. In any case, its stage of develop- ment in the line of true tapirs postulated elsewhere in this paper would appear to be later than Uinta B. Recognition of the amynodont re- mains as Amynodon advenus by Wood in both the Douglass and Bad- water collections would suggest a near equivalence in time. As to Hyrachyus douglassi, it would not appear to be as late as upper Uintan. H. douglassi and Chasmotheroides, cf. intermedius may well be Uinta B, or even earlier. There remains consideration of the faunas from the Swift Current Creek beds of Saskatchewan and the Tapo Ranch horizon of the Sespe in California. Although the collections known from the Swift Current Creek beds consist of decidedly fragmentary materials there is sug- gestion of an age which might not be far removed from that at Bad- water. Contributing to this is the association of lagomorph and Hyopsodus seen in both assemblages. Of the horizons represented in the Sespe sequence, the Badwater would appear to be nearest to that represented at Tapo Ranch or C.I.T. locality 180. Although the species and most of the genera are not the same, the ages are probably not too different. The distinctive nature of the Tapo Ranch fauna may be largely due to its geographic remoteness, 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 132 SYSTEMATIC DESCRIPTION OF THE MAMMALIA MARSUPIALIA? DIDELPHIDAE? PERATHERIUM?, sp. The isolated lower molar, Amherst No. 10019, which A. E. Wood (1949) regarded as questionably representing Metacodon does not seem to me to be insectivore. His careful drawing of this tooth sug- gests possibly a closer relationship to the marsupials. I am particularly impressed by the lingual position of the hypoconulid, and the posterior deflection of the crest on which this cusp is located, away from the entoconid. The talonid appears to be quite different from the structure seen in Jctops and is unlike that, for example, in M, of Metacodon mellingeri where the hypoconulid is closely connected to the entoconid. For these reasons I have tentatively assigned this specimen to Pera- therium?, sp. LAGOMORPHA LEPORIDAE MYTONOLAGUS WYOMINGENSIS Wood, 1949 Plate 1, figure 1 A. E. Wood (1949) has described several isolated teeth of Mytonolagus from the Badwater area and a comparatively unworn P® was designated the type of Mytonolagus wyomingensis. A right maxilla (U.S.N.M. No. 21090) with P*-M? collected by F. L. Pearce undoubtedly represents the same species. P* in this specimen, however, is more worn than in the type, although the teeth in general appear to be less worn than in the type of Mytonolagus petersoni which Burke (1934) described from Uinta C at Myton pocket. The teeth are strik- ingly like those in the type of M. petersoni, but it is noted that the hypostriae on M, and particularly M, are more persistent, extending nearly to the upper limit of the enamel lingually. The comparative weakness of the lingual fold toward the root of M* in M. peterson was further noted in upper-tooth material of the Myton form in the collections of the National Museum. At least the hypostria extends nowhere near the upper limit of the enamel. Wood has regarded M. qwyomingensis as perhaps more primitive than M. petersont. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 9 RODENTIA Description of the known Rodentia in the Badwater fauna has been covered by A. E. Wood (1949). The collection described by him is at Amherst College. CARNIVORA LIMNOCYONIDAE LIMNOCYON?, sp. A maxillary fragment (U.S.N.M. No. 21088) with only P* may represent Limnocyon, but this is not certain. The specimen shows the infraorbital foramen immediately above and anterointernal to the anteroexternal root of P*, much as observed in Limnocyon. The tooth would appear to be a trifle smaller than in Limnocyon douglassi to judge by Peterson’s (1919) illustration of this form. The Badwater tooth measures 9.6 mm. long by 9.4 transversely to base of enamel on the deuterocone. MIACIDAE MIACIS, cf. ROBUSTUS (Peterson), 1919 Plate 1, figure 2 A comparatively large miacid is represented by a lower jaw exhibit- ing the teeth P, to Mz inclusive. Miacis would appear to be indicated by the distinctly basined form of the relatively small talonid of M.. The talonid of M, may likewise have been basined, although most of the superior surface of this portion of the carnasial is damaged so that its precise form is uncertain. It is, nevertheless, as in M., short and decidedly narrower than the trigonid. Mz; is missing, repre- sented by a single alveolus. From measurements given by Peterson (1919), the type of Miacis robustus from the Uinta at Myton pocket is a little larger than the Badwater specimen. Peterson regarded P, and M, as subequal in size so that P, is evidently both relatively and actually larger in the type. This tooth, however, in U.S.N.M. No. 21087 closely resembles that of the type in the presence of a prominent anterior cusp and a strong talonid cusp, followed posteriorly by a well-developed cingulum. The abbreviation of the talonid in both M, and M, likewise suggests M. robustus. The type of Miacis uintensis Osborn (1895) from Uinta B would appear from the scale of Osborn’s illustration to be a little shorter IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 in length of cheek teeth and distinctly shallower jawed than the Bad- water form. Moreover, P, in the type of Osborn’s species apparently lacks the anterior cusp but has one more cusp on the posterior crest, to judge by the illustration (fig. 2), resembling more closely the re- ferred tooth, A.M. No. 1895. M; is relatively smaller, and M, in Osborn’s type has a higher trigonid and a less distinctly basined talonid. Miacis gracilis Clark (1939) is, of course, a decidedly smaller species and P, is evidently characterized by two posterior cusps in ad- dition to the cingulum. MEASUREMENTS IN MILLIMETERS OF LOWER TEETH IN Miacis, cr. robustus, U.S.N.M. NO. 21087 P,, anteroposterior diameter: transverse diameter...............0000. 10.5: 5.4 Mu,, anteroposterior diameter: transverse diameter of trigonid......... T3.05°7:5 M2, anteroposterior diameter: transverse diameter of trigonid......... 5.6: 4.3 CONDYLARTHRA HYOPSODONTIDAE HYOPSODUS, cf. UINTENSIS Osborn, 1902 Plate 1, figure 3 A single Hyopsodus upper molar, U.S.N.M. No. 21089, may well represent H. uintensis, although some doubt may be entertained as comparisons involving such limited material cannot be entirely satis- factory. The tooth is about intermediate in size between M* and M? in the type, A.M. No. 2079, but resembles M? more closely than M?. The Badwater molar measures 4.3 mm. long by 5.6 transversely. This is too small to occlude properly with the type lower molar of Hyopso- dus fastigatus Russell and Wickenden (1933) from the Canadian Swift Current Creek beds. It is interesting to note that although the type, and presumably the two referred lower jaws mentioned by Osborn, are from the Uinta C of Utah, there is in the collections of the U. S. National Museum a lower jaw from Uinta B at White River pocket. Mention may also be made of an upper Eocene occurrence of Hyopsodus at the Beaver Divide. The specimen, an upper molar, comparable in size to the Badwater tooth, was collected by Van Houten in beds he early regarded as representing the Beaver Divide conglome- rate. The locality in question is some distance away from the critical Wagonbed Springs section and Van Houten has since doubted * the 8 Personal communication. No. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN Tr correlation so that the tooth may well have originated in the Uinta equivalent present in the sequence. PERISSODACTYLA EQUIDAE EPIHIPPUS, cf. GRACILIS (Marsh), 1871 Plate 1, figure 5 The rather scant material representing Epihippus was first en- countered in the Badwater localities in 1953. A maxillary fragment, U.S.N.M. No. 21092, including P?, P*, and part of P* and a single lower molariform tooth, U.S.N.M. No. 21094, possibly Mz, represent an equid approximately the size of Epihippus gracilis. P? in No. 21092 is advanced over Orohippus in the development of the lingual portion, but not nearly so molariform as in Mesohippus. The anterointernal cusp in this tooth appears weaker than in the type of Epihippus parvus as figured by Granger (1908), being scarcely more than a low crest, extending lingually from the lingual surface of the paracone rather than from a position anterior to the paracone. There is no evidence of a mesostyle on P?. P* would appear to be en- tirely molariform. The second premolar measures 6.7 mm. long by 5.8 transversely. The lower molar, in comparison with Uintan horses, shows little of diagnostic importance other than size which is close to that of the preserved molar (M,) in the type of Epihippus uintensis (Marsh), placed by both Marsh and Granger in synonymy with E. gracilis. The tooth is a little smaller, though scarcely if any more brachydont than Epihippus (Duchesnehippus) intermedius. The V-shaped crests of the lower molar, however, are a little less acute than in the Duchesnean horse. The metaconid and metastylid are separate at the apex but this has been noted in molars as well as premolars of both the Uintan and Duchesnean Epihippus. The tooth measures 9.0 mm. long by 6.3 wide. EPIHIPPUS, cf. PARVUS Granger, 1908 Plate 1, figure 6 The material of a smaller horse in the Badwater fauna likewise in- cludes a maxillary portion with P? and P*, U.S.N.M. No. 21091, and an isolated lower molariform tooth, U.S.N.M. No. 21093. There is, in addition, the greater part of an isolated molariform upper cheek tooth. I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 P? in the specimen compared with £. parvus is distinctly more pro- gressive looking than in No. 21092 helieved close to E. gracilis. The anterointernal cusp is clearly defined on a crest extending postero- lingually from a position anterior to the lingual surface of the para- cone, somewhat as it appears in the type of E. parvus, but with the long diameter of the cusps directed a little more transversely than in the latter, so that the anterior portion of the tooth seems broader. Moreover, the outer wall shows evidence of an incipient mesostyle, better developed in the type, but no trace of which was found in P? of No. 21092. P? in No. 21091 measures 6.1 mm. long by 5.1 wide transversely ; P* is 6.7 by 6.9. The lower molariform tooth is quite like that (No. 21094) com- pared to E. gracilis but distinctly smaller. It measures 7.8 mm. long by 6.3 wide, comparing favorably in length, but a little broader than molariform lower teeth in A.M. No. 2066 referred to E. parvus. BRONTOTHERIIDAE A fragmentary maxillary portion without teeth but showing root portions of the canine and first two premolars would appear to be of a titanothere. Speculation as to the genus represented would be un- warranted. Enamel fragments of large teeth in the collection may also be titanothere, but this is uncertain as they might equally well repre- sent an amynodont rhinoceros. The Badwater specimen cited by Wood, Seton, and Hares (1936) as Telmatherium, cf. cultridens is half of a lower molar which W. K. Gregory, in a note to Wood, observed, “Granger and I found this tooth to be close to M, of referred specimens of Telmatherium cultridens.” There is, of course, a close resemblance; nevertheless, from the very fragmentary nature of the specimens it is extremely doubtful if among the several genera of Uintan titanotheres all can be excluded from consideration. The specimen from Lysite Mountain identified by G. E. Lewis (in Tourtelot, 1948) as Telmatherium, cf. cultridens I have not seen. It could not be located in the collections at Yale Pea- body Museum. CHALICOTHERIIDAE EOMOROPUS ANARSIUS,* new species ‘ Plate 2, figures 1-3 Type.—Greater part of left side of skull and left ramus of mandible, U.S.N.M. No. 25097. * Anarsios (Gr.), incongruous, strange—in allusion to the large and unex- pected canine. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 3 Horizon and locality—Hendiy Ranch member of Tepee Trail for- mation on Dry Creek, SE4 sec. 9, near line between secs. g and 16, T. 39 N., R. 92 W., Wind River Basin, Wyo. Specific characters —Teeth only slightly larger than in Eomoropus amarorum, but skull proportions and depth of lower jaw much greater. Parastyle of upper molars increasingly prominent from M!? to M%, considerably more extended anteroexternally than in Eomoropus an- nectens, and evidently more so than in E. amarorum. Discussion.—One of the more important discoveries in the upper Eocene of the Wind River Basin is the skull and jaw material of the chalicothere, Eomoropus. The specimen (U.S.N.M. No. 21097) con- sists of the left half of the skull and left ramus of mandible, and was found by F. L. Pearce in exposures on an eastern tributary of the east fork of Dry Creek about 20 miles west of the Badwater Creek localities. The deposits here were mapped by Tourtelot ° as the same formation as that exposed along the south side of Badwater Creek and are believed to be the same age. The species represented was earlier (Gazin, 1955, p. 77) thought to be Eomoropus amarorum, but subsequent direct comparison with the type, A.M. No. 5096, would seem to preclude this possibility. E. amarorum was described by Cope (1881) from a specimen consist- ing of the posterior portion of a skull, a lower jaw, and certain other portions of the skeleton illustrated by Osborn (1913), and derived from the Washakie Basin. According to Osborn, Cope’s specimen probably came from near the base of Washakie B, or the upper Washakie. I suspect that the horizon represented is from higher in the Washakie than suggested, inasmuch as E. amarorum would appear to be more progressive than Uinta B Eomoropus annectens. Comparison of Eomoropus anarsius with the type of E. amarorum shows similarities in the orbital region but the depth of the face below the lower margin of the orbit is conspicuously greater, also the post- orbital process of the frontal appears less prominent and overhanging. The lateral view of the squamosal is similar in the two with the rela- tive position of the external auditory meatus with respect to the glenoid surface much the same. However, the distance between the last molar and the glenoid surface is about 30 percent greater in E. anarsius. Moreover, the depth of the lower jaw is also nearly 30 percent greater and the masseteric fossa n. e deeply impressed and better defined. The two animals would appear to be at about the same stage of maturity, with E. amarorum possibly a little older, to judge °U.S.G.S. Oil and Gas Investigations Map OM 124, sheet 1. 14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 by wear of M, (compare height of cusps shown in lateral view, pl. 2, fig. I, with Osborn’s fig. 3A on p. 267, 1913). Unlike the type of E. amarorum, the cranial portion of the £. anarsius skull is poorly preserved, but fortunately the side of the rostrum, missing in the former, is present in the Dry Creek specimen. Perhaps the most striking feature to be revealed by the E. anarsius rostrum is the enlarged canine, evidently not included in the dentition of Moropus. Absence of this tooth has been generally regarded as characterizing the family although in certain forms the complete formula is not known. The upper premolars were not preserved in MEASUREMENTS IN MILLIMETERS OF DENTITION IN TYPE OF Eomoropus anarsius, U.S.N.M, NO. 21097 Length of upper molar series, M*-M* inclusive, parallel to tooth row...... 54.4 M,, anteroposterior diameter perpendicular to anterior margin........... 14.9 M,, greatest transverse diameter across parastyle and protocone.......... 10.7 M2, anteroposterior diameter perpendicular to anterior margin............ 19.0 M2, greatest transverse diameter across pafastyle and protocone.......... 23'5" Ms, anteroposterior diameter perpendicular to anterior margin........... 19.6 Mz, greatest transverse diameter across parastyle and protocone......... 25.0 Length of preserved lower cheek tooth series, P:-Ms, inclusive.......... 78.0 Bencth, Of fOwer PECMIOIALS bac bre. sia mie 3 o:k’eimipisinrnjas A einbs einai otwin thin are inion 24.0 Reneti-or lower molars: Ma-Ms i. c.cicet ss tame cea cena ne vows tauren ney ee 54.0 P;, anteroposterior diameter : transverse diameter of talonid.......... 12.8:7.8 P,, anteroposterior diameter : transverse diameter of talonid.......... 11.2: 8.1 M,, anteroposterior diameter: transverse diameter of talonid......... 13.9: 9.0 M2, anteroposterior diameter: transverse diameter of trigonid........ 16.9: 10.5 Ms, anteroposterior diameter : transverse diameter of talonid......... 24.0: 10.5 @ Approximate. No. 21097, but the three molars are complete. Eomoropus upper molars are characterized by a lophoid protocone and hypocone, more elongate than in Moropus, and with a distinct protoconule. The ex- ternal wall exhibits a strikingly developed parastyle and a prominent, anteroposteriorly compressed mesostyle and paracone rib. The rib on the metacone is weak or wanting. M* exhibits a spurlike metacone directed nearly at right angles to the similarly developed mesostyle. Comparison with upper teeth in Cope’s type is limited to the lingual portion of M® and no significant differences are observed. Neverthe- less, the anteroexternal root for M? and M® can be observed in Cope’s specimen, and its position is not nearly so forward and outward as in E. anarsius, suggesting rather less extension of the parastyle in the molars. Comparison with the excellent upper cheek tooth series pre- served in the type of Eomoropus annectens, which Peterson (1919) no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 15 described from the Uinta B horizon, shows that the Dry Creek speci- men has molars rather similar, except that the parastyles are strikingly more outstanding and the teeth are about 20 percent larger. Moreover, the rib on the paracone, in keeping with the parastyle, is better de- veloped. The lower teeth of FE. anarsius are quite similar to those in F. amarorum and about the same size, although the anterior premolars may be relatively a little larger. E. amarorwm includes all the lower cheek teeth from P, to Ms, inclusive. In No. 21097 only Pz. of this series is missing, although the inner walls of M. and M; are not com- plete. As Osborn has shown, these teeth are much like those in Moropus; however, in the earlier form Pz» is a relatively larger tooth and M; retains a prominent hypoconulid. Moreover, as observed in occlusal view of the Dry Creek specimen, the crista obliqua in all the cheek teeth following P, joins the posterior wall of the trigonid some- what more buccally and lower than in Moropus, so that the W-pattern is not so well developed. HEELALPETIDAE Simpson (1945) included the Helaletidae in the Tapiroidea, an ar- rangement which is distinctly preferable to including it, as Scott (1941) has, in the Rhinocerotoidea. Scott, moreover, included the hyrachyids in the Helaletidae, and although there is much to be said for such a grouping, as the hyrachyids are in many ways intermediate between the tapiroids and rhinocerotids and show certain marked resemblances to Helaletes, nevertheless the family distinction as the Hyrachyidae within the Rhinocerotoidea, as advocated by H. E. Wood, II (1934) and retained by Simpson (1945), may well be more de- sirable. The Eocene tapiroids are structurally a comparatively conservative group, distinguished from one another by relatively small and seem- ingly unimportant differences. Nevertheless, at least two families, the Helaletidae and Isectolophidae, should probably be recognized for North American forms. DESMATOTHERIUM Scott, 1883 Originally described by Scott as coming from the Bridger Eocene, it is now understood that the type of Desmatotherium guyoti came from the Washakie beds (see Granger, 1909, p. 22) and is in all likeli- hood upper Eocene rather than Bridgerian in age. Peterson (1919, p. 127) was evidently in error in citing the locality for this specimen as ‘‘Henry’s Fork, Wyoming.” 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I TAPIRIDAE OLIGOCENE Protapirus Colodon oe ah Randlett DUCHESNEAN | \oonaten pe ’ il 4 A Helaletes se af Parisectolophus ane Sty Bridger C Heloletes © ! Parisectolophus 1 i fu pel Kd ( Parisectolophus Soo eee \ Lysitean H crati Homogalax EO GE NE BRIDGERIAN Tel %Exoct horizon uncertain Fic. 1.—Suggested phylogenetic arrangement of North American Eocene tapiroids. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 17 Scott’s generic comparisons were made largely with Hyrachyus, from which it clearly differs. Iam convinced, however, that a distinctly closer relationship to Helaletes is indicated. The Badwater, Sage Creek, and Washakie materials together provide the evidence showing the sequence from Bridgerian Helaletes through Uintan Desmato- therium to Oligocene Colodon. As a consequence, Desmatotherium should be included in the Helaletidae, as Simpson (1945) has in- dicated, not with Hyrachyus as Peterson (1919) placed it. On the other hand, the subfamily separation of the Colodontinae from the Helaletinae made by Wortman and Earle (1893), which Simpson has preserved, cannot now be reconciled with the sequential arrangement indicated above. DESMATOTHERIUM WOODI,*® new species Plate 2, figure 4 Type.—Right maxilla with P?-M* (P* incomplete), U.S.N.M. No. 20200. Horizon and locality —Hendry Ranch member of Tepee Trail for- mation on south side of Badwater Creek, SW cor., SE} sec. 14, T. 39 N., R. 89 W., Wind River Basin, Wyo. Specific characters—Upper molar teeth approximately 20 percent smaller than in Desmatotherium guyotu Scott or Desmatotherium kayi Hough. Upper premolars smaller than in these species but relative size intermediate between them and closer to D. guyotit. Discussion.—F our incomplete upper dentitions and a number of iso- lated teeth, including some from the lower series, all from the Bad- water Creek localities, represent the species D. woodi. Two of these, part of the type and two upper premolars (part of U.S.N.M. No. 20202) were figured by Hough (1955, pl. 8, figs. 6 and 9) as material referred to the Sage Creek species D. kayi. I have examined all the Sage Creek specimens together with the Badwater material and find there is no overlap in observed size range for each. The type of the Sage Creek species is nearly 20 percent larger than that of the Bad- water form. D. kayi was described as close in size to D. “guyotit” but with smaller premolars. The premolars of D. woodi are a little smaller than in D. kayi, but the ratio of their size to that of the molars is more nearly as in D. guyotit. The principal feature of the upper dentition of Desmatotherium, distinguishing it from Hyrachyus, is the more progressive condition of the premolars. The divided lingual portion of P* and P* gives these 6 Named for H. E. Wood, II, in appreciation of his work on the Hyrachyidae. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 teeth a distinctly more molariform appearance. Resemblance is im- mediately seen to the earlier Helaletes nanus (genotype) which in a similar way is distinguished from Helaletes bodps. Washakie D. guyotu is a much larger form than Helaletes nanus, but D. woodi is nearly intermediate. Certain of the larger Bridger individuals with progressive premolars, referred to H. nanus, make a close approach to D. woodi in size but the separation of the lingual cusps of P* and P* in any case is clearly not so well effected. Resemblance of Desmatotherium upper teeth to those of Colodon is perhaps even more striking; nevertheless, Colodon can with little doubt be defended as distinct from Desmatotherium. The premolars of Colodon, particularly P*, would appear to be more progressive and, as shown in illustrations given by Scott (1941, pl. 81) of Colodon occi- dentalis, the posterior upper premolars, noticeably P*, would appear to have better defined, more clearly separated, transverse lophs. More- over, a comparison of measurements shows that although D. guyotiu is comparable to Colodon occidentalis in size, the latter has distinctly wider teeth both in the premolar and molar series. This is perhaps most noticeable in the appearance of the posterior loph of the anterior molars which is decidedly longer in the illustration of Colodon. Lower teeth of Desmatotherium are rather poorly represented, except in the Sage Creek collections. They are not known for D. guyotu and only certain isolated teeth and tooth fragments are included in the materials of D. woodi. Characters of the lower teeth of D. kayi were briefly discussed by Hough but somewhat further description, particularly a comparison with the earlier Helaletes, seems indicated. Lower premolars of Desmatotherium in comparison with those of Helaletes are noticeably shortened anteriorly and relatively broad. Particularly significant is the distinctly larger entoconid which in Colodon is quite as large as the hypoconid. There is little evidence for an entoconid in lower premolars of Hyrachyus. The progressive de- velopment of the entoconid toward Colodon, and shortening of the trigonid, give the premolars a more molariform appearance, but the entoconid remains distinct from the hypoconid so that a completely lophoid posterior crest as in the molars is never reached. The lower molars of D. kayi, as in Helaletes and unlike Hyrachyus, show clear-cut transverse lophs with only a very subdued crista be- tween them, a tooth form already realized in Heptodon. The para- stylid, particularly on M3, seems more reduced in Desmatotherium than in Helaletes and much more reduced than in Hyrachyus. The hypoconulid on Ms; may be slightly more reduced than in either Helaletes or Colodon. The lower molars of Colodon, in addition to no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 19 their relatively greater width, are more nearly symmetrical bilaterally, exhibiting a slight crest forward from both the entoconid and meta- conid, quite matching those of the labial side. As previously noted, there seems little doubt but that a phyletic sequence is indicated from Helaletes through Desmatotherium to MEASUREMENTS IN MILLIMETERS OF UPPER DENTITIONS oF Desmatotherium woodi U.S.N.M. No. U.S.N.M. U.S.N.M. 20200 No. No. Type 20201 20202 Length of upper premolar series, P’-P*, in- LEAS TV Cmevetsieee eros syera. si eee cektreferexa ee orat yetaberen Masa ararbce ahah doa tera hehe 33.3 P’, anteroposterior diameter : greatest trans- VERSE GIAITICLSE. fsck c Nehe a eteha Na aceite bie cha PITTS Shak hed 7.5:6.0 P*, anteroposterior diameter.............0. eee CAA Wiest, aoe 8.3:— P*, anteroposterior diameter: greatest trans- MECH COIAMIGLER Oice sche. <'p ljaisiatyaleta stoi elie oes 9.0: 11.7 8.9: 11.8 P*, anteroposterior diameter: greatest trans- MCCS MIAINELER se Sccstaes sence te 8.8: 12.4 0.2::12.7 9.1: 12.6 U.S.N.M No. 20204 Length of upper molar series, parallel to EQOEHE TOW, slonie feiiciawran sateine one seers aaa (WAN cheese Malo achene tate M,, anteroposterior diameter perpendicular ei EAE TAS IO. 6.) ce 16:6 M*, anteroposterior diameter : transverse di- PEPER stele wicen, elena isis cists wore lera éjea lm esate 13-72 19.0 14.6°2 19.5", 15.07 18:3 Length of lower cheek tooth series, P: (at BIVEOIUS )— Dis MICINSIVE. Sos cbs fee fale cite d's MEL fees. eats ec 82.5 Length of lower premolar series, P: (at AIVERNUS = Eay THCMUSIVE ia ciciarcitjarsitjainiaeoieiele ees 43.2 38.3 Length of lower molar series, M:-Ms, in- MIS VEL Meade eta cbse cbere (ein $e» cherace fe ahs dpsed siete ahr pea abee'’s hint over 44.0 Px, anteroposterior diameter (at alveolus) : PLCALESE LLATISVERSE Ciametety se cee aie ocrlts cakeicletah ec elaisietelererdls 10.0: 8.3 P,, anteroposterior diameter : greatest trans- VEUSE WOIAIICLER, jsaraa ti ethe etan tiers crecosteys cifarersi els avb ave 10.0: 5.3 90.5: 4.6 P;, anteroposterior diameter : greatest trans- MEESEMOIAIMELEEN, Ware eaataue Peccrere bis os Gisieiavuelenierere is ET0;:'7.0 10.8 : 6.2 P,, anteroposterior diameter : greatest trans- WEMGE TGATNCLEE? Fisica cence ciate ceeds Vsetaae ss 12.2: 8.0 1153 7.2 M,, anteroposterior diameter : transverse di- AINETEDHOL \talONIG'. 2:22 sc/ers seve oes viaetena betaine Lesieshtanctae 10.9: 8.1 M;, anteroposterior diameter : transverse di- AMeLereOe talOnid’ s,.../aed adale sjagidcee aps da welace TAA 1-3" “1F0Fs0.3 Ms,, anteroposterior diameter : transverse di- AMMELEL TOL THISOMIM kaki siclerdewhactiede! ete sane we —:10.7 20.3: 10.4 2 Approximate. * Measurements of posterior upper premolars are taken anteroposteriorly across outer portion and transversely perpendicular to outer margin. Those of upper molars are taken anteroposteriorly perpendicular to anterior margin and transversely across anterior portion of tooth, lingually to base of enamel or cingulum. 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 toreodon pearcet size is no longer an aid. Teeth of Diplobunops are slightly less selenodont. The inner and outer crests of the upper molars are seemingly farther apart with the outer cusps a trifle more erect and their apices more buccal or not so lingually directed. The cusps or crests of the lower molars have a slightly more inflated or obtuse look and may be relatively broader. There is, of course, variation in both forms ; moreover, these characters of the teeth are scarcely of generic significance and seem somewhat distinctive only between contempo- raries in these groups in the higher horizons of Uintan time. Diplobunops, so far as known, did not progress much beyond the Protoreodon tooth structure characterizing Uinta B time. The genera are, as I have attempted to show earlier (1955), very closely related and their skulls may be distinguished essentially on the different char- acter of the anterior part of the snout. OROMERYCIDAE MALAQUIFERUS TOURTELOTI Gazin, 1955 Plate 3, figure 2 The type of this form is from the Dry Creek occurrence some 20 miles to the west of the Badwater Creek localities. Its description and illustration were included in an earlier paper (1955, pl. 16), and need not be repeated here. An isolated oromerycid upper molar (U.S.N.M. No. 21102) in the Badwater collection exhibits the same rectangular form, rugosity, posteriorly bifurcate protocone, and outstanding ribs on the outer cusps as in Malaquiferous tourteloti. It corresponds closely in form and size, and in the possession of a slightly outward- deflected metastyle to M*. It differs in having a heavier, antero- posteriorly developed mesostyle. The shape of this tooth is entirely different, although structurally related to that in Oromeryx plicatus. Likewise, it would not be confused with the comparatively large Protylopus? annectens. This tooth lends support to the belief that the Dry Creek occurrence is probably equivalent in time to that at Badwater, a conclusion earlier reached on the basis of lithology, and the general upper Eocene indica- tion by the presence of Eomoropus. LEPTOMERYCIDAE LEPTOTRAGULUS, cf. MEDIUS Peterson, 1919 Plate 3, figure 5 Two lower jaw portions, and probably several of the isolated molars, represent a form close or identical to the upper Uintan Leptotragulus no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 31 medius. One of the jaw portions (U.S.N.M. No. 21104) with P, and M, falls within the size range of the Myton material, but the other (U.S.N.M. No. 21103), which has P;-M,, is a little larger than any in the above series. Nevertheless the latter is closer in size of teeth to L. medius than it is to L. proavus. The structure of the lower pre- molars in these jaws corresponds very closely to that regarded as characterizing Leptotragulus. The teeth in both Nos. 21103 and 21104 are significantly smaller than in the type of Leptotragulus? significans Russell from the Kish- enehn beds in British Columbia. Although I have not examined the Kishenehn specimen, from Russell’s clear description and stereo- scopic illustrations I would favor referring his form to Leptomeryx rather than Leptotragulus. As a consequence, there would seem to be somewhat better evidence for the Oligocene age postulated by Russell as an alternate possibility for the Canadian occurrence. LEPTOREODON?, sp. Plate 3, figure 3 A couple of isolated premolars in the collection, a little smaller than in Leptoreodon marshi but not greatly different in size from Lepto- tragulus medius, exhibit a well-defined metaconid. In one of these the metaconid is opposite the protoconid and shows a well-defined groove between them anteriorly, much as in characteristic material of Leptoreodon. The parastylid, however, is a well-developed column distinct from the anterior crest of the protoconid, suggestive of Leptomeryx. Nevertheless, as in Leptoreodon and unlike Leptomery., the talonid basin is formed by the posteroexternal crest swinging lingually near its posterior extremity, joined only by a weak spur from the metaconid. In Leptomeryx, the hypoconid and entoconid in P, are in most cases sharply separated and join forward with the protoconid and metaconid respectively. The second isolated P, ex- hibits a parastylid much as in Leptotragulus, but the metaconid, though weak, is distinct and unlike Leptotragulus. This tooth has a very primitive look and may not represent any known leptotragulids. REFERENCES Burke, JoHN J. 1934. Mytonolagus, a new leporine genus from the Uinta Eocene series in Utah. Ann. Carnegie Mus., vol. 23, pp. 390-420, pl. 50. CLarK, JOHN. 1939. Miacis gracilis, a new carnivore from the Uinta Eocene. Ann. Car- negie Mus., vol. 27, pp. 349-370, figs. 1-2, pls. 34-37. 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 Corr, Epwarp D. 1881. The systematic arrangement of the order Perissodactyla. Proc. Amer. Philos. Soc., vol. 19, pp..377-401, figs. I, 2. Gazin, C. Lewts. 1955. A review of the upper Eocene Artiodactyla of North America. Smith- sonian Misc. Coll., vol. 128, No. 8, pp. 1-96, charts 1-2, pls. 1-18. GRANGER, WALTER. 1908. A revision of American Eocene horses. Bull. Amer. Mus. Nat. Hist., vol. 24, pp. 221-264, figs. 1-5, pls. 15-18. 1909. Faunal horizons of the Washakie formation of southern Wyoming. Bull. Amer. Mus. Nat. Hist., vol. 26, art. 3, pp. 13-23. HatTcHER, JOHN B. 1896. Recent and fossil tapirs. Amer. Journ. Sci., ser. 4, vol. 1, pp. 161-180, figs. I-2, pls. 2-5. HovucH, JEAN R. 1955. An upper Eocene fauna from the Sage Creek area, Beaverhead County, Montana. Journ. Paleont., vol. 29, No. 1, pp. 22-36, figs. 1-3, pls. 7-8. Kay, J. LeRoy. 1953. Faunal lists of vertebrates from Uinta Basin, Utah. Guide Book, 5th Ann. Field Conf., Soc. Vert. Paleont. in N.E. Utah, pp. 20-24. Lewis, G. Epwarp. 1947. Tertiary geology of northwestern Wind River Basin, Wyoming (ab- stract). Journ. Washington Acad. Sci., vol. 37, p. 371. Love, Joun D. 1939. Geology along the southern margin of the Absaroka Range, Wy- oming. Geol. Soc. Amer. Spec. Pap. No. 20, pp. 1-134, figs. 1-3, pls. I-17. MarsH, OTHNIEL C. 1871. Notice of some new fossil mammals from the Tertiary formation. Amer. Journ. Sci. and Arts, vol. 2, pp. 35-44. 1875. Notice of new Tertiary mammals—IV. Amer. Journ. Sci. and Arts, vol. 9, pp. 239-250. MatrHew, WILLIAM D. 1909. Faunal lists of the Tertiary Mammalia of the West. U. S. Geol. Surv. Bull. 361, pp. 91-138. Osszorn, Henry F. 1895. Fossil mammals of the Uinta Basin. Expedition of 1894. Bull. Amer. Mus. Nat. Hist., vol. 7, pp. 71-105, figs. 1-17. 1902. American Eocene Primates, and the supposed rodent family, Mixo- dectidae. Bull. Amer. Mus. Nat. Hist., vol. 16, art. 17, pp. 169-214, figs. I-40. 1913. Eomoropus, an American Eocene chalicothere. Bull. Amer. Mus. Nat. Hist., vol. 32, pp. 261-274, figs. 1-9. PETERSON, Oar A. 1919. Report upon the material discovered in the upper Eocene of the Uinta basin by Earl Douglas in the years 1908-1900, and by O. A. Peterson in 1912. Ann. Carnegie Mus., vol. 12, pp. 40-168, figs. I-19, pls. 34-47. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 33 1931. New species from the Oligocene of the Uinta. Ann. Carnegie Mus., vol. 21, pp. 61-78, figs. 1-12, pl. 1. 1934. List of species and description of new material from the Duchesne River Oligocene, Uinta Basin, Utah. Ann. Carnegie Mus., vol. 23, PP. 373-380, figs. 1-8. RusseELL, Loris S. 1954a. The Eocene-Oligocene transition as a time of major orogeny in west- ern North America. Trans. Roy. Soc. Canada, ser. 3, vol. 48, sec. 4, pp. 65-69. 1954b. The mammalian fauna of the Kishenehn formation, southeastern British Columbia. Ann. Rep. Nat. Mus. Canada (1952-1953), Bull. 132, pp. 92-111, figs. 1-8, pls. 1-3. Russet, Loris S., and WICKENDEN, R. T. D. 1933. An upper Eocene vertebrate fauna from Saskatchewan. Trans. Roy. Soc. Canada, ser. 3, vol. 27, sec. 4, pp. 53-65, fig. 1, pl. 1. Scott, WILLIAM B. 1883. On Desmatotherium and Dilophodon, two new Eocene lophiodonts. Contr. E. M. Mus. Geol. and Archaeol. Princeton College, Bull. No. 3, pt. 4, pp. 46-53, pl. 8. 1941. The mammalian fauna of the White River Oligocene. Trans. Amer. Philos. Soc., vol. 28, pt. 5. Perissodactyla, pp. 747-980, figs. 137-157, pls. 79-100. 1945. The Mammalia of the Duchesne River Oligocene. Trans. Amer. Philos. Soc., vol. 34, pt. 3, pp. 209-253, pls. 1-8. Simpson, Georce G. 1945. The principles of classification and a classification of mammals. Bull. Amer. Mus. Nat. Hist., vol. 85, pp. i-xvi, 1-350. TourTELOT, Harry A. 1946. Tertiary stratigraphy in the northeastern part of the Wind River Basin, Wyoming. U. S. Geol. Surv. Oil and Gas Invest. Prelim. Chart 22. 1948. Tertiary rocks in the northeastern part of the Wind River Basin, Wyoming. Guide Book, 3rd Ann. Field Conf., Soc. Vert. Paleont. in S.E. Wyoming, pp. 53-67. 1953. Geology of the Badwater area. U. S. Geol. Surv. Oil and Gas. Invest. Map OM124 (2 sheets). Woop, Apert FE. 1949. Small mammals from the uppermost Eocene (Duchesnean) near Badwater, Wyoming. Journ. Paleont., vol. 23, No. 5, pp. 556-565, figs. 1-24. Woop, Horace E., II. 1934. Revision of the Hyrachyidae. Bull. Amer. Mus. Nat. Hist., vol. 67, pp. 181-295, figs. 1-51, pls. 20-24. Woon, Horace E., II; Seton, Henry; and Hares, CHArLes J. 1936. New data on the Eocene of the Wind River Basin, Wyoming (ab- stract). Proc. Geol. Soc. Amer. for 1935, pp. 394-395. WortMAn, Jacos L., and EarLe, CHARLES. 1893. Ancestors of the tapir from the lower Miocene of Dakota. Bull. Amer. Mus. Nat. Hist., vol. 5, pp. 159-180, figs. 1-7. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I EXPLANATION OF PLATES PLATE I LAGOMORPH, CARNIVORE, CONDYLARTH, AND PERISSODACTYLS FROM THE BADWATER UPPER EOCENE Fic. 1. Mytonolagus wyomingensis A. E. Wood: Right maxilla (U.S.N.M. No. 21090), occlusal view of teeth. 4Xnatural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 2. Miacis, cf. robustus (Peterson): Left ramus of mandible (U.S.N.M. No. 21087), occlusal and lateral views. Natural size. Badwater upper Eo- cene, Wind River Basin, Wyo. Fic. 3. Hyopsodus, cf. uintensis Osborn: Right upper molar (U.S.N.M. No. 21089), occlusal view. 4 Xnatural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 4. Epitriplopus?, sp.: Left lower molar (U.S.N.M. No. 21099), occlusal view. Natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 5. Epihippus, cf. gracilis (Marsh): Right maxilla (U.S.N.M. No. 21092), occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 6. Epihippus, cf. parvus Granger: Right maxilla (U.S.N.M. No. 21091), occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fics. 7, 8. Dilophodon, cf. leotanus (Peterson): 6, Right upper cheek teeth (U.S.N.M. No. 21098), occlusal view (incomplete P* reversed from left side). 7, Right lower cheek teeth (U.S.N.M. No. 20207), occlusal view. One and one-half times natural size. Badwater upper Eocene, Wind River Basin, Wyo. PLATE 2 PERISSODACTYLS FROM THE DRY CREEK AND BADWATER UPPER EOCENE Fics. 1-3. Eomoropus anarsius, new species: 1, Skull and left ramus of mandi- ble (U.S.N.M. No. 21097), type specimen, lateral view. Two-fifths natural size. 2, Left upper molars (U.S.N.M. No. 21097), type specimen, occlusal view. Natural size. 3, Left lower cheek teeth (U.S.N.M. No. 21097), type specimen, occlusal view. Natural size. Dry Creek upper Eocene, Wind River Basin, Wyo. Fic. 4. Desmatotherium woodi, new species: Composite right upper cheek tooth series; premolars (U.S.N.M. No. 20202), occlusal view; molars (U.S.N.M. No. 20200), type specimen, occlusal view. Natural size. Bad- water upper Eocene, Wind River Basin, Wyo. PLATE 3 ARTIODACTYLS FROM THE BADWATER UPPER EOCENE Fic. 1. Apriculus praeteritus, new genus and species: Right maxilla (U.S.N.M. No. 21100), type specimen, occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. no. 8 MAMMALIAN FAUNA, BADWATER AREA—GAZIN 35 Fic. 2. Malaquiferus tourteloti Gazin: Right upper molar (U.S.N.M. No. 21102), occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 3. Leptoreodon?, sp.: Left Ps (U.S.N.M. No. 21105), occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 4. Protoreodon, cf. petersoni Gazin: Right maxilla (U.S.N.M. No. 211or), occlusal view. Twice natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 5. Leptotragulus, cf. medius Peterson: Right ramus of mandible (U.S.N.M. No. 21103), occlusal and lateral views. Twice natural size and natural size, respectively. Badwater upper Eocene, Wind River Basin, Wyo. Fic. 6. Diplobunops, cf. matthewi Peterson: Right upper cheek tooth series (U.S.N.M. No. 20304), occlusal view (P? and P* restored from left side). Natural size. Badwater upper Eocene, Wind River Basin, Wyo. Fics. 7, 8. Protoreodon pearcei, new species: 7, Right upper cheek teeth (U.S.N.M. No. 20305), type specimen, occlusal view. Natural size. 8, Right lower cheek teeth (U.S.N.M. No. 20305), type specimen, occlusal view (P; reversed from left side). Natural size, Badwater upper Eocene, Wind River Basin, Wyo. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131, NO. 8, PL. 1 LAGOMORPH, CARNIVORE, CONDYLARTH, AND PERISSODACTYLS FROM THE BADWATER UPPER EOCENE (See explanation of plates at end of text.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOEScISi5 INO. 8) Pln2 PERISSODACTYLS FROM THE DRY CREEK AND BADWATER UPPER EOCENE (See explanation of plates at end of text.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131, NO. 8, PL. 3 ARTIODACTYLS FROM THE BADWATER UPPER EOCENE (See explanation of plates at end of text.) a ~ SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 9 3REEDING AND OTHER HABITS OF CASQUED JORNBILLS (BYCANISTES SUBCYLINDRICUS) (Wir 6 PLaTEs) By LAWRENCE KILHAM Bethesda, Md. Steed, *% |}. A te Pi . é REF HSON OF" ° See HINGTOW Se C2000 00?* (PusticaTion 4259) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION NOVEMBER 8, 1956 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U. S. A. PREFACE I went to Uganda at the invitation of the East African High Com- mission to carry on virus research as a visiting scientist at the Virus Research Institute, Entebbe, where I worked from August 1954 until mid-May 1955. My ornithological observations were made as an ama- teur in the early mornings and evenings, and on weekends. It had been my hope to study some particular field problem in addition to making a general acquaintance with African bird life. The nature of the prob- lem was determined soon after my arrival. In my bird notes, black- and-white casqued hornbills [Bycanistes subcylindricus (Sclater) | soon took up more pages than any other species. They came to our garden frequently. In addition, a pair of them roosted and carried on courtship activities in a tree above our house. When I discovered a concentration of hornbill nests in the Mpanga Research Forest, it was apparent that I had an unusual opportunity to study the natural history of casqued hornbills. Present studies did not begin until many females were already walled in. A few pairs of late-nesting hornbills, however, enabled me to witness the beginning stages of nesting ac- tivity. Observations on 16 nesting pairs gave, in the aggregate, a rounded picture of breeding and other habits of these birds. As far as I am aware, this is the first detailed description published on the natural history of Bycanistes subcylindricus. Moreau (1936), how- ever, has written of a related species, Bycanistes brevis. His account is based on the histories of two nests that he observed in Usambara, Tanganyika. Acknowledgments.—The writer is grateful to the following indi- viduals for help contributed in various ways: Dr. A. J. Haddow, director, and Dr. W. H. R. Lumsden, assistant director, of the East African Virus Research Institute ; H. C. Dawkins, ecologist, Uganda Forest Department; Charles Sandison, curator, Botanical Gardens, Entebbe, for identification of fruits and seeds; Dr. V. G. L. Van Someren, Ngong, Kenya, for identifying insect remains; and Dr. Herbert Friedmann, curator, division of birds, U. S. National Mu- seum, for aid and encouragement in preparation of the manuscript. Two sketches of hornbills by their nests were contributed by my wife, Jane Kilham. The avian scientific nomenclature used is that given by Mackworth-Praed and Grant, 1952. LK. iii CONTENTS Page eT Ce eee epee ea oy tres Sahat RG eee eve aT aT aT et bnb celal rac ohies Shovel a eseelie ayocabe\wis a's iil MEIER OG UCTIONS «art ccyrersis eis er atsk isso sister ctoveserehete rer ave) eVoid susie ayers areatels © sieraverets)e.sleve ie I IRS E EE OS Beate ciara zicte lw ceeiahere are STe Levee he Ste Eo sla eue) shaves aie) Sielaharetwrel'ara vole o'2 o.bleia) 2 2 Generalvhabitstorehornbills areas oe rareeietoie wateke aici al clare eiaimiaie le ered slevaleleisiore 4 ESE CONSEDUCE Of i cpcinis clare araie erat Sroferera her siets avatars isiforeiatel scaimlarel adit die\ejessisieve ina erat 8 PNGHIVITIES s Ole TCSEITIS PAILS cca clofnh cxe/etstelsveraravetste. ciel ctelers Mines isles! ** *°Or CCazeet sr re Gy PoaLI-ci °° Pr vS-Z -z1"* yS-b -zI°° vS-P -zI*" ¥S-Iz-I1"* bS-0z-I1"" bS-tz-11°* PS-PI-I1°* PS-VI-I1"* ¥S-9 -II°° vVS-p -II°° $S-of-o1°" bS-bz-o1°** ¢S-z-01'* pe19a090stp aeq “ET “er ry ‘OI 6 “S tery ‘ON 3S9N SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 132, NO. 9, PL. 2 2. Young male hornbill 2 days after emerging from nest 5. 1. “Mpanga,” hand-reared male hornbill 10 months of age. “Zika.’ hand-reared female hornbill 10 months of age. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131, NO. 9, PL 1. Location (shown by arrow) of nest 13 in stub of broken limb. >, Nest 5 (location shown by arrows) in Piptadenia tree in Botanical Gardens, lentebbe. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOE.°237, INO 9) Pia 1. Termite mound at headquarters clearing, Mpanga Research Forest, where female hornbill collected earth. 2. Entrance of nest 10, showing cement wall NO. 9 CASQUED HORNBILLS—-KILHAM il ideally located. It was 70 feet up in the trunk of a huge tree (Celtis saoyouxit), as illustrated by a sketch (text figure 2). Unfortunately the hole was about a foot in diameter and apparently too large. The sketch shows the small wall built across the lower portion. Its rate of construction was exceedingly slow, even though the female worked for N Fic. 1.—Nest 3, Mpanga Forest. many hours, usually in the morning. On October 24 the pair flew to the hole at 8:45 a.m. The female went inside and the male perched close by. He offered her a mud pellet 5 or 6 times, but she was occu- pied and paid no attention. At 9 a.m. the male flew to the forest clearing, then down to a path among the shambas. Here he picked two gobs of damp earth, swallowing the first and holding the second, an inch in diameter, in his bill. Then he flew back in stages to the nest tree. I watched him cough up and pass three small mud pellets 12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 to his mate. She took them at her bill tip in rapid succession. The male then remained quietly by for some time while she continued working inside. Twenty minutes later the male again flew to the for- est clearing, alighting in a patch of maize. I crept up to within 30 feet and found him perched on a stump several feet above the ground. Fic. 2.—Nest 1, Mpanga Forest, never completed. He was bending over repeatedly, and I could see that he was picking up lumps of earth and swallowing them. To my surprise the female flew over. Both birds now perched together at the forest edge while he coughed up and transferred 5 or 6 pellets to the bill of his mate. Then he hopped away a few feet, broke off a 2-inch piece of bark, and bounced back to offer it to her. She refused it. When the pair had flown into the forest with loud squawking, I examined the stump. A termite nest clung to the side of it. Freshly opened tunnels, now NO. 9 CASQUED HORNBILLS—KILHAM 13 lined by soldiers, showed where the hornbill had collected earth. On October 27 the male again visited the stump. At 12:30 noon he flew from his nest tree to the maize patch where I was able to observe him from a distance of 25 feet by using a screen of corn stalks. This time he picked up considerable earth from the stump and a little from the ground. Then he flew back to his nest hole, where he spent some time perched on the rim, moving his head about inside where his mate was working. Both birds were silent. So far it appeared that his job was to gather dirt and hers to build with it. However, at 8:20 a.m. on October 31 both birds flew to the forest clearing. She flew to the ground and hopped behind a mud-wattle hut. I moved around to see her bounce up against the wall and knock off a piece of dry mud the size of a plum, which she swallowed. She next hopped around a corner and knocked off another piece. From here she flew up to rejoin her mate. Shortly afterward she followed him to a distant pawpaw tree, where he fed her some fruit. Within 15 minutes they were back at the nest. By 9:10 a.m. the female was working on the nest wall, making a rapid tapping noise like a woodpecker. The male flew away for a short time, apparently to collect dirt. On his return, he perched beside his mate outside the hole, then gave her 11 pellets which she accepted at the tip of her bill and swallowed. She did not use them immediately, for the pair flew away from the nest hole and were gone for a half hour or more. These observations showed that female hornbills as well as males collect dirt, although they probably do so less often. The female from nest 1 did not always use pellets immediately after swallowing them. Dirt might be retained in her gullet for half an hour or more. The same was true of the male. Since earth was generally hard and dry when collected, the period of retention would give it a chance to be- come moistened by glandular secretions and by juices from such fruit aS paw-paws retained in the gullet at the same time. A larger part of the moistening may be done by the male. I noticed with my captive hornbills that the male could swallow more than twice as much as the female. He is, of course, a third larger in size. It was of interest that hornbills came to termite nests to gather building material, as these nests are exceedingly durable; each grain of earth is selected and coated with a cement substance by the termites. I saw the birds from nest I visit a termite nest on another occasion. The pair flew to the edge of the forest clearing at 9:08 a.m. on November 7. The female flew to a low tree, then over the shambas to the tall termite mound shown on plate 4, figure 1. Clinging to its steep sides, she took up 15 chunks of dirt and swallowed them. Finally her mate flew over to 14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 join her. Developments which took place in the next few minutes were unexpected. Coition——The pair had been nest building for at least 2 weeks when they perched together on the termite mound. There was a touching of bills, and the male acted as though he were trying to give her a few extra pellets, but he apparently had none to give. At 9:15 a.m. the female flew from the termite nest, closely followed by the male who made a loud, wailing noise. The two headed toward the nest tree, but lighted on a branch at the forest edge. I had to run across the clearing for a better view. Within this short time the male had mounted the female. He came off a few moments later. Then he mounted on her back again, without hurry, and got securely placed. There was no noise. She had her tail held way over to one side. When he mounted he pressed his tail downward and somewhat under her body. His wings were kept closed. After some moments he got off and flew alone to the rim of the nest. She continued to hold her tail to one side. I could see against a background of white feathers that her vent was extroverted but was retracted shortly afterward. From at least 9:45 a.m. until 12:20 p.m. the female apparently worked from inside the nest on the wall, as she had on previous mornings. Gathering of dirt at other nests—At 5 p.m. on November 7 I was watching nest construction by a pair of hornbills in the Botanical Gardens. The male flew toward me and lighted on the ground 35 feet away. Here he picked up 15 or more chunks of earth. After he had flown away, I examined the spot which proved to be a low termite mound. Freshly opened tunnels showed where the bird had been working. Although hornbills appeared to be especially attracted to termite nests they are not invariably so. Sometimes they pick up ordinary dirt. On October 31 the male from nest 4 lighted on a stick close to the ground of a native shamba. He repeatedly bent over, took up chunks of black cultivated soil, and swallowed them. His next move was to an adjacent banana tree. Here he tore off pieces of leaf and bark, 1 to 2 inches long, three of which he swallowed. His subse- quent flight to the forest led to the discovery of his nest hole. The wall of this nest was unusually black. Construction of the nest wall—At various times I watched four pairs of hornbills constructing their nest walls. There was variation in the time of day when work took place. Three pairs worked in the morning and one pair, from the Botanical Gardens, late in the after- noon. Experience at nest 2 was typical. On October 24, the male returned to his tall, isolated nest stump at noon. He leaned over the nest opening as he heaved up pea-sized pellets of dirt. These he NO. 9 CASQUED HORNBILLS—KILHAM 15 passed with his bill tip to that of his mate directly below. At least 10 pellets were passed in rapid succession. The male then bounced along to another part of the stump and tried to whack off bits of dead wood. In a few moments he bounced back to the hole. He shook his head from side to side, with bill half open, as though a pellet had gotten stuck. Then for a while he held his head low over the hole to watch what was going on. A week later I found him carrying on much the same. He spent considerable time looking down and even putting his head and neck through the hole. Sometimes his mate accepted two pellets and refused a third. At such periods of active construction, the only sound was the rapid tapping of the female’s bill, which could be heard some distance away. Male hornbills did no work on the walls, although they might peck and explore about a nest entrance. A male usually sat by like a brick- layer’s helper. He would fetch building material and supply it to his mate as needed. The pair at nest 1 would spend the larger part of a morning in this fashion, settling down to work at about 8:30 a.m. Work at the Botanical Garden nest was done from midafternoon on. At this time the sun shone directly into the nest entrance. Possibly this pair was taking advantage of the illumination. I could see the female’s bill tapping inside. Details of how the tapping was done could not be perceived readily in high nests. I had a better opportunity at nest 7 which was comparatively low. The female was already walled in but on January 29 she was repairing the entrance with material from the nest floor. She applied this material with rapid tapping of the side of her bill tip. Tapping was again seen to good advantage in my captive pair of hornbills. In September 1955, when 9 months of age, both birds would get in a corner of their cage and plaster the wall with any litter, feces, or food matter which might serve as mud. Both sexes did the tapping. This activity was carried on daily, as judged by the appearance of the wall and the amount of tapping we heard. Abortive nesting—On November 7, following coition, the pair of hornbills from nest 1 had worked all morning on the nest wall. By the following weekend their interest had obviously declined. The hole was too large to wall in. On the afternoon of November 13 the male kept returning to the nest rim. Then he would fly back to the forest to join his mate. Sometimes the pair would fly together to the nest tree, making a tremendous noise. It was a dismal wailing. The male, however, would look into the nest alone. On the following day I observed the same behavior from 7:15 a.m. until 12:30 p.m. These two were the noisiest hornbills in the forest. It appeared that he was 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 anxious to continue with the nest, but that she had lost interest, as during the whole morning she never came near it. He would put half of his body inside and make low grunts as he poked about with his bill. The same performance continued on a following weekend. She entered the nest several times, started tapping, then flew out again after a short time. The male persisted for two more weeks in his efforts to have her return. December 4 was the last time I ever saw her in the nest. This was 6 weeks after I had first found her at work. My final view of the pair was on December 12. They were perching near the nest cavity but showed no interest in it. Structure of wall and the inside of the nest.—Nest entrances were usually elongate slits, 14 to 2 inches wide and 4 to Io inches high, depending on the size of the natural cavity. These measurements are approximate. Plate 4, figure 2, a photograph of nest 10, shows a representative nest entrance. Walls were remarkably strong. Neither Africans nor myself, by putting a hand into the slit and pulling hard, could budge or loosen them in three nests opened for removal of young. Much hacking and prying with a curved bush knife were needed to effect an entrance. Another indication of the stoutness of walls was provided at the time of natural nest openings. One-half of the entrance cement of 2 nests (3 and 12) was knocked out entire, apparently by the emerging female. These pieces fell 60 to 85 feet to the ground, where I picked them up unbroken. They were roughly 4 inches long, 3 inches in width and thickness, and were built in con- centric layers. Possibly each layer represented a day’s work. Odd bits of insects, bark, and plant material were incorporated in the cement in haphazard fashion. In two of the three pieces, one side was dark brown. It did not run with the concentric layers and it had faced the inside of the nest. Females had probably made this addition after being walled in. For lack of earth they had used fruit stones, seeds, and what appeared to be darker fecal material. It did not appear that feces was an integral part of the main cement structure in any of the five nest walls I examined closely. There was nothing organized about the inside of a nest. Nests 10 and 14, examined at time of opening for removal of young, had frag- ments of rotten wood at the bottom. Nest 10 also contained many large contour feathers. When I pulled away the cement from this nest, hundreds of small ants swarmed out from behind the edges. My hand and arm were covered with them as I explored the cavity which was roughly a foot in diameter. An African who broke open nest 14 for me had a similar experience. These ants are presumably attracted by feces and other debris, for an amazing amount of fruit NO. 9 CASQUED HORNBILLS—KILHAM 17 appears to go right through a hornbill’s intestinal tract undigested. This is especially true of young birds, as I discovered with my captive specimens. Feces from younger individuals does not always clear the entrance. This was observed from the scaffold by nest 10. There was, therefore, ample organic matter to attract ants, and their presence in great numbers may explain why I failed to find insects, in the nature of parasites, scavengers, or others, in examinations of nest debris. ACTIVITIES OF NESTING PAIRS Walling in of female.—As far as I can determine, I was watching nest 5 on the evening when the female began her 4 months of volun- tary confinement. On November 7, the male was active late in the afternoon bringing termite earth to his mate, and giving her pellets as she worked. On the following day I arrived at 5:15 p.m. to find him bending over the entrance. The sun shone into the hole. I could see that the nest wall was complete and that no more building was in progress. For the next hour the male stayed close by. At 6:30 p.m. he flew to a large tree (Canarium schweinfurthi) and picked some fruit, then returned to his nest with loud wailing. He gave his mate five or six fruits and she made low guttural notes. Dusk was now coming on. The male flew to a limb 15 feet away, then took a long flight which carried him out of sight over the brow of the hill. His mate remained alone. It was not until March 7 that I was to see her again. Feeding and other attentions of male to mate in nest—Once walled in the nest, the female was entirely dependent on the male for food until she emerged with her young one some months later. Nest 5 was well situated for observations. I would station myself below the huge Piptadenia tree in the Botanical Gardens (pl. 3, fig. 2) every few days during the week, before breakfast and again late in the afternoon. The male was accustomed to people passing beneath the arching limb containing the nest hole. Some days I would wait nearby an hour without seeing him approach the nest. I discovered, however, that there was some regularity to his visits, one usually taking place close to 7:30 am. Forty feeding visits were observed in the course of 4 months. A visit on November 13 was typical. At 5:40 p.m. the male came flying over the open lawns, head held high, and wings making a “woo-ooshing” noise. He held a chip of wood 5 inches long in his beak. It looked as though he were flying with a cigar. He landed on the big limb, then bounced sideways until he was on the bole above the nest entrance. Then he leaned over and pushed the 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 stick of wood through the entrance. When his mate had taken it, he coughed up eight blackish fruits in succession, swinging his head down each time to place them in the tip of her bill. By the time the feeding was done, the stick had come out of the hole. He picked it up and pushed it in again. Then he flew to a perch 15 feet away. At such times he would usually whack his bill loudly back and forth on a limb as one would whet a knife. This male, like most of the others I observed, did not linger in the nest tree after a feeding visit. Stick or bark presentation was a common prelude to feeding. It occurred in 13 of the 40 visits observed. Objects presented ranged from curled pieces of bark, 1 by 6 inches, to smaller bits an inch in diameter. Discarded and dropped pieces accumulated on the ground below the nest. The male was persistent about these offerings. On December 8 he lighted above the nest and swung his bill down into the entrance 23 times to offer a piece of bark. His mate gave no re- sponse. When she accepted on the twenty-fourth try, he fed her four fruits. The male usually made a feeding chuckle when a fruit was trans- ferred. Numbers of fruits offered varied from 2 to 17 per visit, but counting was often difficult. Thus on February 7 the male fed his mate 17 small fruits. Some of these were offered 4 or 5 times before she accepted. On February 28 he fed her 2 “cherries.” A third one, however, had to be held down 11 times before she took it. Possibly the female is at times occupied with the young chick, so that she is not in a position to accept. Fruits brought to the nest ranged from the size of a pea to that of a small plum. In addition to bringing sticks and fruit, the male of nest 5 cleared away accumulations from the entrance. The approach to the nest, formed by a broken-off limb, sloped slightly upward from the wall. It thus collected fecal and other matter expelled from the nest. The male lingered to clear away this debris after 14 of 40 feeding visits. Sometimes he swallowed a few items. More often he would pick up small bits with his bill and toss them outward in a rapid and system- atic fashion. On February g he did this 30 times after one visit and on February 28, 25 times. Usually he made only a few tosses before flying away. Nest 7 was the only other nest where I observed a male clearing the entrance. Observations at other nests, while generally similar to those made in the Botanical Gardens, differed with the individual character of hornbill pairs. Sometimes physical peculiarities of the nest cavity led to differences of behavior. Nest 6 was in the straight trunk of a huge tree and the female often rested with her tail protruding from the 130) VOL. SMITHSONIAN MISCELLANEOUS COLLECTIONS Female hornbill at time she was removed from nest 10, approximately I. period. oO d two-thirds of the way through nestin months of age. > when approximately “Zika”’ and a “Mpaneg > NO. 9 CASQUED HORNBILLS—KILHAM 19 entrance. When the male bent over to offer food, she would not always bother to turn around. One day he gave her a fresh green leaf before coughing up four yellow fruits. At nest 12, also at Mpanga, the female would put her whole bill out of the entrance to take food. Possibly females in these last two nests were crowded for space. Some of the holes, such as those of nests 4 and 7, were on straight trunks without boles. Consequently, the visiting males had to cling to the lower rim of the nest with tail fanned out against the trunk. An occasional visiting male would bring bark to his mate but no food. Expulsion of feces, and other activities of female in nest—Watch- ing and listening from the ground gave some insight into activities of nesting females. At infrequent intervals one might see a stream of fecal matter shoot 2 to 3 feet out of an entrance hole, glisten in the sun, then land with a splash on the leaves below. The white feathers of the female’s rear end were, in some nests, clearly visible as she maneuvered her vent to the opening. Hornbill vents are protrusible and mobile. This could be well seen in our young birds 6 to 8 weeks of age when, standing way up on their legs, they would back up over the edge of the box they were in and expel feces on the floor. Expul- sion was not as forceful as seen in some hawks. Tree trunks and foliage below nests were not appreciably stained by expelled feces. This was partly due to the dark color of the droppings resulting from a fruit diet. Some streaking of white appeared in feces with develop- ment of the young. On January 8, as I was sitting on the scaffold be- side nest Io in Mpanga Forest, the 4-weeks-old chick backed to the en- trance and deposited a cylinder of feces 4 inches long on the lower cement. This feces had a white film over one end. The ladder leading to nest 10 became increasingly spattered with feces as weeks went on. Observations on my captive birds indicate that the white substance in the feces increased with ingestion of animal protein. I saw one nesting female toss debris out of the entrance with her bill. Doubtless this method also contributes to nest sanitation. Females within the nest did not lose their constructive instincts. Bill tapping continued, but was carried on far more by some females than others. I frequently heard tapping from nests 7 and 10, which were only 120 feet apart, at the same time. Much of the time I could not see a bill in the aperture. It is conceivable that these birds plaster debris against their nest cavities either because habit is strong or just by way of idle amusement. My young captive hornbills plastered the walls of their cage, possibly for the same reasons. Entrance walls, however, sometimes needed repair. On November 28, the female of nest 6 was repairing her nest entrance at noontime. I could see her 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 bill tapping rapidly on either side of the lower aperture. The shape of this aperture changed somewhat from one week to another and the repaired areas were darker in color. Presumably feces and other debris present by the opening were used, for I later obtained half of the cement from this nest and found that the dark areas had seeds and fruit stones embedded in it. Nesting females may enlarge their nest cavities by pecking at rotten wood surrounding them. The female of nest Io had an escape attic above her nest. I could hear her scuttling into it when I climbed up the ladder, and on looking through the opening all I could see was the tip of her tail. When nest 14 was opened on January 21, there was no female in sight. The African who had removed the chick swept the whole length of his arm inside without encountering the mother bird. His position was too precarious for him to look inside. It seemed probable that the mother had crawled into some remote recess. EGGS, YOUNG, AND NEST OPENINGS Eggs.—Nest 10 was in a dead tree 30 feet above the ground in Mpanga Forest. On December 4 I climbed the scaffold to this nest for the first time and peered through the aperture, using a flashlight. The mother bird was almost out of sight in her escape attic. There were two white eggs, similar to those of a domestic fowl in size and shape. My next visit was on December 11. The forest ranger said that he had climbed to the nest at 6 a.m. and had seen two eggs. I approached the nest tree at 1:30 p.m. and saw two-thirds of an egg- shell on the ground directly below the nest hole. The shell was so fresh that ants were still swarming over its moist inner surface. I climbed the scaffold to find the mother hornbill facing me at the entrance. This was the only time she ever did so. As far as I could determine before she climbed to her escape attic, she had a complete plumage. When she left I saw one egg and one blind, completely naked, rather blue young one. This was a first view of my subsequent pet, Mpanga. When I looked in on the following morning, I could not see him, but he soon emerged from under some debris, giving a feeble ““chirpee, chirpee.” His lower bill was larger than the upper one. Early in the morning of December 14 the ranger found the second egg chipped, and by afternoon he saw the shell on the ground and a second chick in the nest. I was able to visit the nest two days later and see the two chicks together. The larger one was chirping lustily. He had brown mash over his bill and throat, and there was more mash in the nest. I wondered whether the mother hornbill had regurgitated food onto the nest floor and then fed it to her offspring. NO. 9 CASQUED HORNBILLS—KILHAM 2I It was January 1, 1955, before I was able to visit the nest again. There was now a single young one, the size of a plucked pigeon, which begged and peeped a few times when I looked in. The forest ranger reported that the second chick had disappeared a week after hatching. The remaining chick had its eyes open. They were dull but mobile. Pinfeathers were just beginning to emerge on its head and wings. The entrance hole was becoming stained with feces, whereas it had been clean previously. On January 8, the young bird had black pinfeathers one-fourth of an inch long on head, neck, and wings. There were smaller, colorless pinfeathers on back, tail, and in two tracks bordering the breast bone. Feet and an inch-long fleshy tail ap- peared large for the size of the bird. The chick seized my finger when I pushed it in. He also chewed pieces of wood. This bird was removed from its nest when 6 weeks old and has lived well in captivity for over a year. Periodic inspections of the ground below nest trees gave clues as to the number of eggs and approximate time of hatching in four other nests. For example, on November 27 I found pieces amounting to two-thirds of an eggshell, with its membranes, below nest 4. I care- fully removed all pieces, and on November 28 there were no further eggshells. Six days later, however, I found a second eggshell, three- fifths intact, with an additional one-fifth in pieces. It appeared that two eggs had been laid and that they had probably hatched on different days. This had happened at nest 10. There, it may be recalled, the eggshell was tossed from the nest soon after the hatching of the chick. On the ground below nest 9, I removed most of an eggshell on Novem- ber 28, and three-fourths of a shell on December 4. These and other data are tabulated in table 1. Data from two other nests were less complete. I found over half of an eggshell below nest 3 on November 28, but it was not until January 1 that a second shell turned up. Pos- sibly it had been covered with debris, either in the nest or on the forest floor. On December 9 there were pieces of one shell below nest 12. I did not, as in the other cases, know how long they had been there. Breaking open nests to obtain young. Molting of female.—Inacces- sibility made it difficult to study the molt in nesting females, but I was able to gather some information. Nest 10 could be reached by ladders. The female was usually in her escape attic and I did not want to interfere with her by making an opening. The nest was well lined by remiges when I looked in on December 4. In retrospect I should have removed them with a pair of long forceps for arrangement and count- ing. All I could see of the mother bird was her tail. The tail feathers remained soiled with no evidence of renewal. On December 11 I had 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 my only full view of the mother when she briefly defended her chick. Her plumage appeared complete, but I did not see her outstretched wings. My next view of her was when I broke open the nest on January 22. After putting the 6-weeks-old chick in a bag, I reached into the hollow trunk and pulled the mother bird down. She was kept in captivity for a few days of observation. Plate 5, figure 1, shows that her plumage was complete. The only sign of molting was one tail feather, a few inches long, which was still enclosed in a sheath. She was not shy in captivity, but she remained motionless, as if stunned, and refused to eat. I liberated her on January 24. In spite of a long period of confinement in the nest and having had no food for 2 days, she flew to a tree, squawked a few times, then took a flight of 300 to 4oo yards. She was headed back in the direction of Mpanga Forest, 13 miles away. It was apparent that this female, viewed when roughly one-third, and again at two-thirds through the nesting period, had not experi- enced any sudden or complete molt. On January 30, we opened nest 16 to remove a chick 6 or 7 weeks of age. The mother bird struggled vigorously, striking the African who held her a sharp blow on the chin, so that he fell over backward. When I took hold of her it was obvious that she was in no weakened condition. Her plumage ap- peared to be complete except that her tail feathers, although well grown, had sheaths at the base. She flew readily to a tree when liberated. Premature escape of female due to loss of mate-——When I entered Mpanga Forest on the afternoon of January 1, I heard the wailing screech of a female hornbill in distress. The calls were given twice a minute. I followed them to nest 4 where I found a pair of foreign hornbills. These flew away at my approach. The female in the nest kept screeching for the next 2 hours in a most pitiful manner, but her mate failed to return. I examined the ground below the nest tree and found that he had been killed. There were two large patches of feathers directly below his usual perch. These patches were 2} to 4 feet across. One consisted principally of small body feathers and the other of large feathers from wings and tail. I suspected that the hornbill had been struck from his perch by some bird of prey, and, after falling directly to the ground, had been plucked on the spot. The female was still screeching when I left the forest late in the afternoon. On the following morning I reached the nest shortly after 9 a.m. The mud wall was partially broken out. A new pair was inspecting the nest and it was evident that the original female was no longer there. Natural nest openings.—I observed how nests were opened naturally NO. 9 CASQUED HORNBILLS—KILHAM 23 in five nests with entrances visible from the ground. In each the cement had been knocked away from one side of the aperture. This left ample room for the mother and young to emerge. I was interested to find that the missing cement was lying in an intact piece on the ground below three of the five nests. These five nests (Nos. 3, 6, II, 12, and 13, table 1) all opened between January 1 and February 5. Some of these may have been open for a week before I noticed them. Nest 6, however, was closed on the afternoon of January 22 but open by 9:15 a.m. on the following day. There were no hornbills in the vicinity. Emergence of mother and young.—On February 5 I noticed that the female of nest 7, which I had had under observation for 84 days, was still walled in. By the following morning she had left. I began to search the adjacent forest and was able to locate the pair 100 or more feet from the nest tree. The female was recognizable by her soiled plumage, the white patches of which were muddy. The male, recognizable by his bill markings, sat close by her. For the next 2 hours I hunted back and forth through uncut jungle. It was raining hard and I thought that if I could find the young, I could probably catch it if its plumage was water soaked. The parents expressed great alarm, coming down within 20 feet of my head. Unfortunately I could not find the young. I wondered if it had crawled into some hollow limb. I had better success in the Botanical Gardens. The female was walled in on November 8, 1954, and had emerged with her young one on March 7, 1955. She was confined for 119 days, with a possible error of 2 days. There were no signs of activity by the nest on the morning of March 7. Late in the afternoon, however, I found the male perched by his mate 50 yards from the nest tree. He made continuous noises. Several times he hung his head way back, allowing her to nibble the feathers of his throat. Her plumage was in poor con- dition. The white parts of her feathers were soiled, her tail rumpled, and the small feathers on the back of her head and neck were worn. There was no sign of the young one. At 7:15 a.m. on March 8 I located the pair by cries coming from a patch of forest. They were together in a tall tree, and a young hornbill was close by. His plumage was in fine condition, pure black and white, his tail nearly full length, and his upper bill had the large, light-colored growing patch of a male (pl. 1, fig. 2). He made squawks similar to those of my captive birds. Everything appeared well when I left. The tragedy that overtook the family during the morning may be reconstructed from the chance observations of another bird watcher, 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 Mrs, Iris Darnton. At 10:30 a.m. Mrs. Darnton was attempting to photograph the parent hornbills where I had seen them earlier. The young one was perching on a low branch by a roadway. He flew with some difficulty to a higher perch. At this moment an intruding female hornbill attacked the young one and the two fell grappling to the ground. The parents made a great commotion. Their young one lay flat on the road, but soon flew onto the lawn, then into a low tree. After 5 p.m. I came to the gardens and found the family where last seen by Mrs. Darnton. The young bird was perched precariously near the top of a spindly tree and one foot hung limp and useless. He was using half-spread wings to maintain his perch. The male parent made a great noise when he saw me, but soon quieted down, hopped closer, and fed the young one four fruits. Ten minutes later he tried to feed him again, but without success. The mother bird did not attempt to feed the young one. She remained inactive. A foreign female hornbill stayed about 50 feet away. On the following morning I found the parents in the same area, but the young bird was not in the trees, so I searched the underbrush and found him on the ground. When placed on the lawn, he was unable to fly. The male parent swooped repeatedly at my head. I was reluctant to take away the young bird, but it was obvious that any passing dog or individual could kill it. I therefore took it home. Plate 1, figure 2, shows his appearance 2 days after leaving the nest. I estimated that he was Io to 11 weeks of age, using as a guide my captive male of known age. One of his feet was broken. When placed in a splint, it healed completely in 3 weeks. This bird was the only one of my four captives that did not become tame. Parental devotion.—No large hornbills had come to our garden regularly since departure of the roosting pair in October. However, from April 1 until May 15, when we left, a pair of hornbills came every day, often remaining for some time. I soon recognized the male. It was the one I had watched for 4 months in the Botanical Gardens, which were 2 miles away. The parent hornbills had located their young in his outdoor cage, and our garden resounded off and on all day with their wailing and commotion. They perched on adjacent trees, and frequently swooped down low over the wire. “Mutesa,” as we called the young one, never responded in any perceptible way. TERRITORY, AND RELATIONS OF HORNBILLS WITH ONE ANOTHER Specific interference-—Nesting hornbills were interfered with by members of their own species to a surprising degree. Experiences at nest 5 illustrate the persistence of such interference. On November 6 NO. 9 CASQUED HORNBILLS—KILHAM 25 the pair of hornbills were working on their nest late in the afternoon. An adult male kept coming into the tree and the male in possession repeatedly drove him away. By November 8 the female was walled in, and a more serious attempt at interference was now made by a foreign female. I first noticed her on November 19. She was follow- ing the male and lighted in the nest tree when he lighted above his nest hole. On November 23 the same course of events took place, except that the male was less tolerant. He fed his own mate, then drove the intruder away. A week later I again saw her fly in close behind the male and light 25 feet from the nest hole. The male gave his mate a piece of bark followed by some fruit, and then bounced from one branch to another toward the foreign female. The intruder called and the female within the nest screamed a number of times. I wondered whether the interloper could seduce the male, but from subsequent observations it seemed unlikely that she would. The male returned again to the nest hole, and a few minutes later was in the upper part of the tree knocking about on dead branches until he dis- lodged a piece of bark. He clamped his bill on the bark until it was largely fragmented. Then he moved toward the foreign female. If he presented the bark, one would suppose that she had some attraction for him. After a moment, however, he changed his direction, flew down to the big limb below, bent over the nest hole, and gave the token to his mate, accompanied by a feeding chuckle. Subsequently he returned to perch quietly within 8 feet of the intruding female. At 7:30 a.m. the two of them flew away together. As the nesting season progressed, he became less tolerant of her intrusions. On my next visit, a week later, he made several swoops in an effort to drive her away, but she was not discouraged. On February 3 I again watched her fly in behind the male and alight in the nest tree, making considerable noise. The male stopped feeding his mate, swooped at the interloper and drove her down toward the ground. However, when he flew away, she followed a short distance behind. It generally appeared that her interest was in him rather than in the nest. On March 2 I observed a more serious situation. Late in the afternoon I found a foreign female clinging to the nest entrance. This time she was alone. She worked industriously, removing debris from the en- trance and knocking from the cement wall chunks which she broke in her bill before dropping them. There was no noise. After 5 minutes the male arrived and she flew a short distance away. He tossed some debris from the entrance, then drove the foreign female to another tree, flying at her so hard that he knocked leaves from intervening branches. He returned to his nest with a small stick held like a cigar. 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 His mate, who had remained silent, now began her wailing screeches. I also heard her bill tapping. The intruding female, persistent as usual, had followed the male back to the nest tree. In a few minutes he flew at her again, flying faster than hornbills usually do as he chased her from one tree to another. Five days later, mother and young emerged from the nest. As already related, a foreign female attacked the young bird and apparently broke its foot. After I had picked up the helpless young one on March 9g, I returned to the Botanical Gardens late in the afternoon. The pair of hornbills were perched side by side in their nest tree. Not long after I heard a great flutter of wings. I looked back to see both members of the pair pur- suing a foreign female. This was the last I saw of her. When the parents later came to our garden, she did not follow. I have presumed that the same foreign female was involved in all these incidents relating to nest 5. This presumption was based on her consistent behavior, general appearance, and bill shape. I never saw another female with which to compare her near the tree. At 5 p.m. on January 26 I witnessed an intrusion by a pair of hornbills. A foreign female was on the lower rim of the nest entrance, poking her bill about the aperture. She made no noise. After some minutes a foreign male lighted on a limb above. He had a fruit in his bill tip. The female moved toward him, took the fruit, and kept offering it down inside the hole. It was not accepted. The foreign female would toss the fruit about in her bill, then try again. Finally the rightful male returned, drove the intruding pair away, and fed his mate a number of fruits. The whole incident appeared odd. I wondered whether the foreign pair were unsuccessful nesters, who, having a strong, though thwarted instinct to feed something, dropped in on the female in nest 5. Interference by foreign hornbills was not limited to the nest in the Botanical Gardens. It happened not infrequently at other nests. A pair of hornbills were involved in each of the following intrusions. On November 19 a foreign pair were perched by nest 4 in Mpanga Forest. The female flew to the entrance, clinging to the lower rim with tail outspread for support. She then gave some hard pecks against the mud wall and grappled at bill point with the female inside the nest. Neither bird made any noise. However, when the intruder withdrew, the nest owner rattled her bill in the entrance. The foreign male sat quietly by without participating. In a period of Io minutes the intruding female attacked the nest entrance 12 times, but did no significant damage. In the next 5 minutes she attacked only twice. Then the rightful male returned and drove the trespassers away. NO. 9 CASQUED HORNBILLS—KILHAM 27 Since this episode took place early in the nesting period, I conjectured that the foreign pair had, perhaps, not found a suitable nesting site and the female was trying to take possession of one already occupied. A second episode was difficult to interpret. It took place late in the nesting period, on January 23. I saw a foreign pair fly into the tree containing nest 15. The male repeatedly bent over the nest rim and there was a rattling of bills. He produced a “cherry” at his bill tip. Then he either gave it to the nesting female or dropped it into her nest. The female rattled her bill at the strange male. Ten minutes later the foreign female swung dramatically on a long tangle of epiphytic roots, then landed on the nest rim. This was the only time I ever saw a pair of foreign hornbills perched together on a nest. The intruding female waggled her bill vigorously in the opening. A few minutes later the owning male swooped in and drove the in- truders away. He fed his mate some yellow fruit. She now screamed repeatedly. On one occasion I saw a lone male attacking a nest. This was on November 21 at nest 6 in Mpanga Forest. The foreign male came quietly to a limb above the nest, then dropped to the nest rim. He appeared wary, bending his head to one side, then to the other, as he hung his head down to look through the entrance. The female had her bill ready but made no noise. He finally struck at the cement, then sparred with the female within the nest through the opening. I could hear their bills clashing. After 12 minutes the returning male owner drove the trespasser away. He had a leaf in his bill tip which he gave to his mate along with some fruit. It should be mentioned that a male may attack his own nest. At midmorning on November 28 the male from nest 7 flew down and rattled his bill in the opening. He was apparently in a bellicose mood for he next flew to a limb directly over my head, which was unusual. I could see his bill markings clearly. Meanwhile, his mate rattled her bill in the entrance. Lone females were the most frequent intruders at hornbill nests. I often saw one at nest 11. She had a favorite perch 10 feet from the entrance where she would sit for some periods. The nesting female would rattle her bill and scream repeatedly, but her mate, on feeding visits, paid little attention to the intruder. On December 12 a foreign female perched calmly on the bole above nest 6. She repeat- edly leaned down into the opening, giving the feeding chuckle eight times as she did so. The female in the nest rattled her bill. After I5 minutes the male returned and drove the foreign female away. A final and most unusual case of interference occurred at nest 16, which was 30 feet above the ground. On January 30 we had placed some 28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 ladders and an African was preparing to climb up and open the nest. As we stood below, a foreign female flew in and lighted on the edge of the nest. I thought that the mother had already emerged, but it was later discovered that the mother and young were inside. Territory —Neighboring pairs of hornbills tolerated each other well. For example, nests 7 and 10 in Mpanga Forest were within 120 feet of each other and a third pair made persistent attempts to nest within an equal distance. Within this triangle I could watch activities of all three pairs at one time. They paid no attention to each other, their sense of territory apparently being limited to the nest tree and its immediate vicinity. Some trespassing was accidental and with- out interest in the nest itself. Thus nest 4 was centrally located in Mpanga Forest so hornbills engaged in their various activities fre- quently came near it. The male from this nest did more chasing than any other I had under observation. He had a peculiar habit of perch- ing during the day within 10 feet of his nest hole. No other male perched close to its nest. Some might occasionally perch within 100 feet, but usually I saw males near their nest trees only on feeding visits. Females were frequently alone for 45 minutes to an hour at a time, and during these periods the nest tree was open to intruders. Hornbills may have a sense of territory in relation to roosting areas ; I did not make sufficient observations to determine whether this was so. Aggregations of hornbills —Sometimes a number of hornbills would come together, usually owing to a common attraction such as a flight of insects, a fruiting tree, or a passing hawk. This did not appear to be true flocking. Nine was the largest number of hornbills I ever saw together. They were in our garden on August 26. Other observers told me that they had seen larger gatherings. On November. 28 I watched two males and four females closely besetting a harrier hawk. A week later I came across what appeared to be the same group in the same area of Mpanga Forest. This time a band of redtail monkeys were working along the forest edge. Six female and two male horn- bills followed along with them sitting in the same trees but making no noise. I believe that the association was an idle one, for the horn- bills were playing with sticks and showed no alarm. The excess of females was of interest. Pairs of hornbills were also frequent during the nesting season. I wondered if there had not been enough nesting sites to go around. On February 15 I saw a curious association of two adult male hornbills. They came to our garden and hunted to- gether closely, going over cracks and crevices in a big tree. Two NO. 9 CASQUED HORNBILLS—KILHAM 29 weeks later I found the same pair a mile away and still closely associated. RELATIONS WITH OTHER BIRDS Hornbills became much disturbed when a hawk or eagle appeared in Mpanga Forest. A crowned hawk eagle perched on a tall tree at the edge of the headquarters clearing. Then he sailed into the forest and was lost from view. An hour later I heard a great noise and found the hawk eagle surrounded by casqued and the smaller pied hornbills (Tockus fasciatus). None came closer than 20 feet. On November 28 I was watching nest 8 when I heard a number of horn- bills making short flights from one perch to another. This drew my attention to a harrier hawk in a dead tree. Two female hornbills were perched within a few feet of him, one on either side. Two males were in the same tree. When the hawk flew, all four hornbills followed him closely but made no noise. The bird that upset hornbills the most was a great sparrow hawk. On December 11 he flew up close to me in Mpanga Forest, calling “ker, kee, kee” in plaintive fashion. Three male hornbills accompanied him. None of them made any noise. Whenever the hawk circled and returned, the hornbills pursued closely and even swooped at him. On January 2 I again heard the cry of the great sparrow hawk. When he lighted above me, a male hornbill lighted within 6 feet of him, and when he flew, two hornbills followed within 20 feet. Hornbills are occasional predators themselves. Their presence, however, seldom caused any disturbance among smaller birds. I saw one hornbill momentarily beset by sunbirds and colies when he was robbing a nest of the latter. Broad-billed rollers (Eurystomus afer) would pursue hornbills passing by the lake shore. These aggressive birds attack everything from anhingas to starlings. FOOD Fruit—Food brought by male hornbills to their nests consisted largely of fruits, ranging in size from a pea to an olive. Some fruits, such as figs and pawpaw, were brought in as amorphous pieces. The elliptical fruits of Canariwm schweinfurthii were conspicuous and prevalence of their stones on the ground were a helpful clue to the location of nest trees. I was able to collect various seeds and fruit stones by cleaning the ground below nests. Following is a list of all fruits identified. Such indigestible matter passes through the digestive tract of the hornbills and is expelled with the feces. This was observed in both wild and captive birds. I have never seen hornbills go near 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I water and my captives do not seem to know what it is. Apparently they get enough water from fruit. FRUITS IDENTIFIED FROM SEEDS, STONES, AND PIPS RECOVERED FROM FECES BELOW HORNBILL NESTS Canarium schweinfurthii Engl. Antiaris toxicaria (Pers.) Lesch. Pycnanthus angolensis (Welw.) Exell. Chlorophora excelsa (Welw.) Benth. Pseudospondias microcarpa (A. Rich.) | Eugenia jambolana Lam. Engl. Dracaena steudneri Engl. Maesopsis eminii Engl. Ficus natalensis Hochst. Animal food.—Bannerman (1953), writing of Bycanistes subcylin- dricus, states that “this bird lives entirely upon fruit, as indeed do most of the large hornbills.” During initial observations I had no reason to doubt this statement. By closer watching, however, I found that hornbills take a wide variety of animal prey. On December 6 a male went from one low perch to another among our garden trees, sometimes only 7 feet from the ground. Five minutes later I saw him fly up from the foot of a jacaranda with a 5-inch lizard squirming in his bill. He flew to a perch over a native shamba. Here he tossed the lizard about in his bill for some time, holding it by the tip of the tail, then chewed along until he reached the head. Finally the hornbill lost hold and the lizard fell. In a feat of acrobatics, the hornbill fell down after his prey, disappearing from sight in the vegetation. Fif- teen minutes later the bird was back in our garden. A completely limp lizard hung from his bill as he flew over the hill in what I suspected was the direction of his nest. On January 31 another male hornbill was hunting in our garden. He hopped onto a perch, looked around slowly in all directions, then hopped to another perch and did the same. After some minutes he flew to a thick bushy tree, where he scrutinized the foliage carefully, then hopped directly to the end of a branch where a mouse bird had its nest. The hornbill picked up a small egg with his bill tip, sent it flying back into his gullet with a toss of his head, then did the same with a second egg. To finish off, he seized some nest material and dropped it. What followed was an example of the delicate control casqued hornbills have with their bills. The male coughed up one egg from his gullet and held it again in his bill tip. By this time his mate had arrived in a tree across the lawn. He flew over to her, still holding the egg, and settling beside her, presented her with the egg; then heaved and presented her with the other, both intact. She swallowed both. On February 15 I watched two male hornbills hunting in our garden. A double-toothed barbet (Lybius bidentatus) was excavating in a tree when the hornbills flew NO. 9 CASQUED HORNBILLS—KILHAM 31 directly to the hole. One male repeatedly drove his bill into the exca- vation. I was not sure whether the barbet was inside or not. For the next 15 minutes the hornbills quietly examined the hole, knocked off dead bark, and searched leaves and seed pods. Their hunting was not successful while in the garden. However, hornbills probably catch other birds on occasion. On January 29 a male in Mpanga Forest glided to his nest with a sparrow-sized bird, chewed beyond recogni- tion, hanging from his bill. He saw me and quickly flew away. My captive hornbills have been maintained to a large extent on left-over laboratory mice which they swallow whole. They appear to thrive on animal protein. Hornbills catch insects both large and small. I found the remains of some insects which they had fed upon by examining fecal matter below nest 5 in the Botanical Gardens. Dr. V. G. L. van Someren was able to identify the following: REMAINS OF INSECTS RECOVERED FROM FECES BELOW HORNBILL NEST 5 Dicranorrhina micans (Drury) Longicornis beetles of cerambycid group Rhyparobia grandis (Sauss.)—large cockroach Long-horn cricket Tenebrionid beetle, Metallonotus Slender-winged mantis Some were not adequate for species identification. Large bark beetles were a frequent finding. Activities of a male hornbill observed on February 1 indicated how these might be captured. For a half hour in midmorning he worked on the dead portion of a large tree. He would strike slanting blows to loosen a piece of bark, then pry under it and appear to pick out something from underneath. The next step was to knock the loose bark to the ground. This hornbill showed acrobatic skill, often leaning way over on its long legs, with head and neck outstretched, in an effort to reach more bark. He removed at least 3 square feet before flying away. Praying-mantis nests were not infrequent in accumulations below nest 5. They had come through the digestive tract more or less intact, as indicated by adherent feces. I would not have supposed that casqued hornbills could catch small insects on the wing. They are, however, quite resourceful in this respect. On April 18 three hornbills were perched on a casuarina tree late in the afternoon. They were gulping at a close swarm of insects that were whirling about in a small cloud near the treetop. These insects were black-winged termites. Some were still moving upward 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 from the ground. The hornbills made a clapping noise as they snapped at the passing termites and were at the game for over 10 minutes, On May 1 I witnessed a similar spectacle, again late in the afternoon. Compact swarms of small insects (not lake flies) were hovering at the tops of tall trees adjacent to the Institute Compound. Smaller birds were catching them, mostly by perching on a topmost spray. These birds included splendid starlings (Lamprocolius splendidus), bulbuls (Pycnonotus tricolor), Abyssinian gonoleks (Lanarius eryth- rogaster), and didric cuckoo (Chrysococcyx caprius). Two casqued hornbills were catching insects along with the smaller birds. For over 20 minutes they kept turning their heads to snap at the swarm around them. Dr. W. H. R. Lumsden has contributed an observation which fur- ther indicates the agility of these large hornbills. On September 6, 1953, he was in the woods of the Botanical Gardens. Three or four hornbills were perched about 60 feet from the ground. They would swoop down across an open space, pick up something in midair, then rise to a perch in an opposite tree. They were after winged ants which were swarming on ground and vegetation below the clearing. SOME ANATOMICAL FEATURES IN RELATION TO FUNCTION Some peculiarities of hornbill anatomy came to have more signifi- cance with continued watching. The large eyes are unusually mobile for a bird. Casqued hornbills can look up and down to a moderate extent without cocking their heads as many birds do. This gives them, by human interpretation, a more intelligent expression. The upper and lower eyelids are continuous and in sleeping this fused eyelid is pulled over the eye from back to front. The combined eyelids are white in adult females. Considering that the eye is dark and sur- rounded by blackish feathers, I have wondered whether these white eyelids enable the male to see his mate better when looking into a dark nest cavity. The head is covered by fluffy feathers, 14 to 2 inches long. These are used in emotional expressions and when fully erect the head is like a small, round feather duster. From front view the topmost feathers, which may be the only ones erected, may resemble two horns. My young captive hornbills demonstrate how these feathers may be used. If I toss grapes to them, Mpanga may grab them all. Zika, the female, then feels left out. This is obvious by her expression. Her head feathers stand straight out in all directions as though to say “Where do I come in?” When alarmed or excited, her head feathers lie tightly back. If she next investigates some object, NO. 9 CASQUED HORNBILLS—KILHAM 33 like a crumpled piece of paper, they stand out partially again. In young birds the feathers just above the eyes and forming the horns are brown. By the age of 10 months these are almost entirely replaced by black feathers. The bill tip can be used as delicately as a pair of forceps. In fe- males, only the tips may be in apposition, leaving a slightly open space for several inches behind. This space is more exaggerated in older females and may suggest, to a mild degree, the bill of an open-bill stork (Anastomus lamelligerus). At first I thought the space was due to wear. However, my captive Zika had this space at an early age when wear was not apparent. The bill tip itself is very sensitive. It is, for example, continually used to investigate strange objects. If I give my captive hornbills a fruit they have not seen before, they will toss and squeeze it in their bill tips for some time before swallowing. The bill of the male is huge, and that of a young one is larger than a female’s before he leaves the nest. This is shown in the photograph (pl. 5, fig. 2) of two young hornbills, male and female, at 7 to 8 weeks of age. Plate 6, figure 1, shows Mpanga’s bill at 6 months of age and plate 2, figure 1, at 10 months. The white patch is apparently an area of growth. With some transillumination one can see that it is full of blood vessels. The patch becomes smaller in older males. No one knows, as far as I am aware, how long it takes the bill of the male to reach full development, with a forward projecting knob. At present I can only speculate on the function of this huge structure. It would appear to have no strictly practical use, for the smaller bill of the female serves more immediate functions. Her bill not only is used to plaster the wall of the nest, but is also a formidable weapon for defending the nest opening. Its inner capacity is no less than that of the male’s. In addition, males and females are equally adept at catching small objects with their bills. I wonder, therefore, if the casqued bill is not chiefly sexual in function. Possibly it is com- parable to the mustache stripe of the male flicker or the red breast of the cock robin. In this sense it would serve to release behavior patterns in the female which promote successful breeding and pair formation. | The feet of hornbills, with three front toes somewhat webbed, do not grasp tightly. I have never felt any real grip from the birds perching on my arm. Hornbills can, however, hang down from a perch, almost parrotlike, without losing their hold. The long tail is remarkable in that it can be neatly folded over the back. This adapta- tion is convenient for females walled within nest cavities, as is the fact that they are a third smaller than males. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 DISCUSSION OF FACTORS CONTROLLING HORNBILL POPULATIONS Birds such as hornbills which rear a single young one are, one may suppose, relatively safe from enemies. This was probably true before the original rain forests had been cut. A female walled in a nest 70 feet above the forest floor, in the trunk of a huge tree without lower branches, is in an almost impregnable position. Such trees, however, are now entirely absent over large areas. Natives’ shambas, elephant grass, and patches of second growth cover the countryside, leaving only thin fringes of large trees along lake shore and swamp. Eastern Uganda is good agricutural country and the native population is rapidly expanding. Interference by man’s activities is, I believe, the greatest factor limiting hornbill populations. Mpanga Research Forest remains as a needed refuge. Even here, observations suggested that suitable nesting sites were way below the demand. Some pairs of hornbills were nesting in unfavorable situations. For example, nests to and 16 (table 1) were only about 30 feet from the ground and were easy to reach. Also, I continually saw pairs of hornbills that were not nesting during the nesting period. Two pairs tried without suc- cess to build nests in unsuitable locations. When nest 4 suddenly became vacant owing to the death of the male, another pair of horn- bills took it over immediately. Some of the incidents of specific inter- ference already narrated indicate the degree of competition. The disastrous effects of forest destruction on casqued hornbills is well described by Capt. C. R. S. Pitman (1955, personal communication). He writes that “ever since I first went to Entebbe in 1925 forest destruction in the vicinity of Entebbe and Kisubi, and in fact all along the NTB-Kampala Road, has been on such an appalling scale, that annually large numbers of trees, with the best nesting sites, are being destroyed. Bycanistes therefore is constantly having to move farther and farther afield to find suitable nesting sites. When I first went to Entebbe there must have been at least two dozen Bycanistes nests within a 2-mile radius . . . but now good nest sites are fewer and far between and Bycanistes resorts to hollows, some readily accessible, which it would have ignored in the past.” Fortunately Africans in eastern Uganda do not molest birds to any extent. Ease of growing food and comparative prosperity probably puts less pressure on them to do so. But in Bwamba, where hornbills were considered fair prey, I continually came across Pygmies and other natives wandering about with slingshots and small bows and arrows. Under these conditions I found the birds more wary and difficult to observe than near Entebbe. NO. 9 CASQUED HORNBILLS—KILHAM 35 COMPARATIVE STUDIES OF OTHER HORNBILLS Genus Tockus.—There were two other species of hornbills in the vicinity of Entebbe, the crowned hornbill (Tockus alboterminatus) and the pied hornbill (Tockus fasciatus). These two smaller horn- bills are somewhat similar in size and appearance. I could never dis- cover any basic difference in their habits. Their high piping cries, erratic type of flight with many rises and dips, and greater concentra- tion on insect food readily distinguished them from casqued hornbills. All three species occurred in the same stretches of open country and forest. On March 20, 1955, I noticed a crowned hornbill flying through Zika Forest. He lighted on a treetop, then suddenly dropped down- ward. Searching the area, I found a leaning tree with a bole, 4o feet above the ground, with a 2-inch hole in the center. There was almost no suggestion of a mud wall. I watched for 20 minutes. At one time white feathers closed the entrance as the female pushed her vent to the opening, and a stream of excreta shot out. The maneuver was the same as I had witnessed with Bycanistes. Later the male returned to perch on the bole and feed his mate a large insect (mantis?). He did not linger, the briefness of his visits apparently being due to the fact that he carried only one item in his bill tip ; there was no heaving up of fruits from the gullet such as characterized visits of male casqued hornbills to their nests. On March 25 an African, well trained at the Institute, climbed up and inspected the nest for me. There were three white eggs. The mother bird, when poked, backed to the rear of the cavity. Unfortunately, preparations for leaving Africa prevented an adequate study of these birds. I am indebted to Dr. Friedmann (1925) for the following account, hitherto unpublished, of the opening of a nest of this species in Kenya Colony. On April 7, at Taveta, some natives cut down a large tree in which there was a nest of this hornbill containing the adult female and two young birds. The nest was about 50 feet up in the tree and was in a large hole, the entrance of which measured roughly 10 inches in long diameter and 3 inches wide. This entrance was plastered up with dry mud, bird feces, and bits of bark all mixed together, leaving an opening about 2 inches long and 1 inch wide. As I picked away at the mud the adult female pecked at me with its bill, about an inch and a half of which could protrude through the opening. When finally I opened the nest and took out the birds I found that the two young birds were of different ages, the older of the two [pl. 6, fig. 2] being feathered on the back, wings, sides, legs, and crown, while the tail feathers were free of their sheaths for their distal thirds and the sheaths of the neck and breast feathers were beginning to burst. The other bird was less well feathered, the wings and flanks being the only parts really well covered. The tail feathers were about the same as in the older bird and the under 36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 tail coverts in both were well developed. In both nestlings the bill showed no trace of a casque and was a yellowish-horn color; the feet were dark lead color; the skin light pinkish; the iris bluish gray. The tail folded up against the back so well as to look like a definite adaptation to living in crowded quarters. In fact it seemed to be muscularly easier for the birds to hold their tails up than to straighten them out [pl. 6, fig. 2]. One of the nestlings when put on the ground fell over forward on its head and breast and the tail remained sticking straight up in the air as though the bird were unable to drop it into what would be con- sidered the normal position. The adult female when about to enter a nest before egg-laying usually begins to molt and is for some time thereafter in quite a help- less condition. New feathers grow in while the eggs are incubating and the young growing to the flying stage. The female taken from the nest had all the new tail feathers well developed but all of them were still basally enclosed in their sheaths. The bird was still missing the outermost secondaries and innermost primaries but the other remiges were there, most of them more or less still in their sheaths basally. The bird could fly only very poorly and seemed dazzled by the light. Several times I let it go and each time it flew or rather half fluttered, half flopped through the air very laboriously for a short distance and stopped by smashing against a tree or the side of my tent. I had some evidence that pied hornbills also breed in March. From November on I had been observing a pair of casqued hornbills in the Botanical Gardens and had kept watch on a squirrel hole 50 feet up in a nearby tree. It was not until March 1 that I noticed a pair of pied hornbills showing any interest. At 7:30 a.m. a pair were preen- ing nearby. Between them they made Io visits to the hole, poking their bills inside. When a crowned hornbill appeared, they chased it away. The following day the pair were at the hole morning and eve- ning. On March 8 I saw them putting their bills into the hole and tossing out debris. I had no subsequent evidence that the pair nested. The hole may have been occupied by a squirrel which I had seen using it previously. Apparently smaller hornbills may compete with hole- nesting mammals. On February 18 I was driving near Kaboona, in the arid country of Karamoja, when I noticed a pair of Jackson’s horn- bills (Tockus jacksoni) catching insects and flying to a 2-inch hole in a dead tree. When I returned 4 days later the pair were still in- specting the nest hole. Thinking young hornbills might be in the tree, I cut it down. To my surprise, the cavity contained a mother bush baby (Galago senegalensis) with a mouse-sized young one. These limited observations may have interest because I could find no breeding dates for these three species of Tockus in eastern Uganda. The investigations of Gordon Ranger (1949-52) offer an opportunity to compare the habits of Bycanistes with those of Tockus in some detail. These investigations on African hornbills are the most com- plete known to me. They have extended over many years and concern NO. 9 CASQUED HORNBILLS—KILHAM 37 another crowned hornbill (Tockus alboterminatus) which occurs in South Africa. Comparisons will be made first in regard to differences of behavior, then to points of similarity with Bycanistes. All observa- tions and quotations on the crowned hornbill are from Ranger’s publications. Differences in behavior between Bycanistes and Tockus.—(a) Crowned hornbills have a definite territory which is fairly extensive, is defended against trespassing hornbills, and is maintained year after year by the same pair which temporarily share it with the offspring of each season. I found little evidence that Bycanistes subcylindricus maintains a definite territory other than the immediate vicinity of the nest tree. (b) Crowned hornbills live more extensively on insects. This greater consumption of animal protein is reflected in their white excreta. In feeding his mate at the nest, the male carries the food, usually a single insect, at his bill tip. He does not load his gullet, then cough up fruits one at a time as do male casqued hornbills. Further- more, crowned hornbills make casts of indigestible seeds, pips, and hard parts of insects. Casqued hornbills, on the other hand, pass every- thing out in the feces—even large fruit stones, whole baby mice, and mantis nests. (c) There are a number of differences in the manner of plastering nest walls. Crowned hornbills make plaster of feces, finer soil from the floor of the nest, and insect remains. According to Ranger “the female does not swallow anything for the purpose of disgorging it in the form of plaster,’ and “the male plays no part whatever in plastering the nest hole.” Bycanistes collect soil and lumps of earth from the ground. Both sexes do this, but the male brings the most and is a kind of “bricklayer’s helper,” supplying his mate who does the actual plastering. (d) A distinction between Bycanistes and Tockus lies in the time of emergence of the female from the nest. Ranger (1955, personal communication) has extensive data on this subject. He has found that the female may emerge 62 to 74 days after being walled in. At this time the precocious young reseal the entrance. Both parents then feed the young which emerge 19 to 34 days later. Moreau (1936) has collected similar information in regard to Tockus deckeni and T. alboterminatus. Similarities in behavior between Bycanistes and Tockus.—A close relationship between the two genera of hornbills is indicated by simi- larities in their behavior patterns. Many of Ranger’s descriptions (1949-52) of the habits of Tockus alboterminatus australis apply 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL: I3t equally well to Bycanistes subcylindricus. I have quoted a number of these verbatim. (a) Food. The crowned hornbill’s handling of larger prey is the same as for Bycanistes. Speaking of a grasshopper, Ranger writes “the hornbill . . . subjects the creature to prolonged chewing and pulping between the mandibles before swallowing, turning it about and tossing it to secure new holds. . . .”” Chameleons and nestling birds are treated in the same manner. Among insects taken by crowned hornbills, those as diverse as winged termites, long-horned beetles (cerambycids) and, curiously enough, mantid egg cases were all fed on by Bycanistes. After feeding, cleaning “is performed by scraping and wiping the outside of the bill against a branch.” (b) Roosts. The crowned hornbill has roosting sites which are used in rotation. Each member of the family, however, has its own private perch. The two casqued hornbills in our garden always used the same individual perch each night. Like Bycanistes, the crowned hornbill is not an early riser. The birds stretch and make gruff utter- ances to each other for some time in the morning before leaving their roosts. (c) Play and agility. The following odd traits are also true of casqued hornbills. Ranger wrote that the crowned hornbill reveals “its dexterity when by diving it recovers an object that falls from its bill before it reaches the ground.” Speaking of a captive bird he writes that ““Conkie was adept at catching objects cast at her over intervals of many yards.” The bill whacking of male casqued horn- bills was a characteristic sound in the forest. Ranger wrote as follows of the crowned hornbill: “The meaning of the exaggerated scraping of the bill against a branch, indulged in more particularly by the male, has not become apparent.” (d) Basking. “The foliage bath is followed by basking, advantage being taken of the sun’s appearance in a clouded sky, but basking is independent in purpose. The body with wings extended is relaxed and spread limply upon branch or foliage, the head and neck upturned. Conkie assumed the most limp, lifeless, unbirdlike attitudes, neck curled with throat uppermost, eyes obscured by the relaxed third eyelid.” Such postures are the ones assumed by my pet hornbill, Zika. It is not a matter of drying her plumage but love of sunshine for its own sake. As soon as the sun comes from behind a cloud, whether she is indoors by a window or outside, Zika assumes the grotesque attitudes so well described by Ranger. (e) Courtship. In describing breeding habits of Bycanistes, I have included various activities under a heading of courtship and main- NO. 9 CASQUED HORNBILLS—KILHAM 39 tenance of the pair bond. This is a matter of interpretation. Ranger uses other phraseology, but the activities he describes are similar. For example, he wrote of the following behavior as having taken place 26 days before final entry. “Investigation of a knocking and rat- tling near Site I disclosed the hornbill pair, one striking its bill with vibratory rapidity against a branch. The side of the terminal part of the bill was used, and the point, vibrating, traveled around the surface of the branch till in turn the opposite side came into play . . . then the other bird .. . became enlivened and extending its bill performed the same rattling action.” Ranger believed this rattling was the same motion employed in plastering and made special note that both sexes were involved. I am not sure whether this perform- ance is entirely related to the onset of nesting in Bycanistes. My captives, Mpanga and Zika, do a good deal of tapping. They began when 9 to 10 months of age and sometimes do it against my clothes. Ranger has stated further that his crowned hornbills made increasing visits to the nest tree as the season progressed. Such flights were initiated by the male. A new behavior was noted 19 days before final entry into the nest when the male began to present food to the female. This was done anywhere, not necessarily near the nest tree. Finally, bark presentation was frequent among crowned hornbills. Ranger found that the female would take bark with ready interest from her mate, then bite it to pieces. (f{) Intruders. I have described intrusions on nesting casqued hornbills by members of their own species. Apparently a similar phenomenon takes place among crowned hornbills. Speaking of a feeding visit Ranger wrote “the male and a young intruder arrived, and this drew a single sharp cry from the female. . . . The male then delivered an item and resumed his chasing of the intruder.” This male subsequently delivered “13 items of food and bark, but all the time was worried by the young trespasser who followed him again and again to the nest. . . .”’ I was unable to tell whether the female in- truders I saw by Bycanistes nests were young birds or not. The male intruders were all adults. Ranger has also written of the nesting female rattling her bill in the entrance. He describes this “habit rattling” as useless activity. This was not true of casqued hornbills. Every time I saw a female rattling her bill there was some cause, such as presence of intruders, to evoke this alarm signal. (g) Plastering. Photographs of nest entrances presented by Ranger show that the cement walls look much the same as those con- structed by Bycanistes. The female crowned hornbill has the same technique of plastering. ‘Always the bill works rapidly in vibratory 40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 fashion, the side of the end portion . . . beating against the surface, to which the moist dung is applied.” Ranger has also noted that the cement walls are built up in layers. Ranger has been fortunate in having many years in which to study crowned hornbills. If I had had at least a second year to study Bycanistes I should have been specially interested in finding out (1) whether these hornbills remain paired from one season to another ; and (2) whether the same pair returns to the same nest tree in suc- ceeding years. Both of these situations, true for the Tockus albotermi- natus, presumably hold for casqued hornbills. Ground hornbills—tI had only casual views of the huge ground hornbills in Karamoja and in Murchison Falls National Park. These form the third main group of hornbills that occur in British East Africa. The following unpublished account of Bucorvus cafer (Schlegel) is contributed by Dr. Friedmann. It is of interest from the point of view of comparative biology. This giant hornbill was seen in rather small numbers in the open bushveldt at Taveta, Kenya Colony, during March and April. The birds were usually seen walking around on the ground in loose groups of three to six individuals. They really walk, not hop. In East Africa they are protected as scavengers and are not molested by big-game hunters and settlers. Although they feed on the ground they sleep high up in tall trees and can fly remarkably well for their bulk. The original “take-off” seems to give them some little difficulty, but when once under way they fly more directly than do most hornbills, their heavy wings causing a very audible woof woof with every stroke. The call note is a deep boom boom, a rather hollow, and reverberating note. During the mating season the birds become more vociferous and call to each other with great frequency. The natives in Kenya Colony have a story to the effect that the female ground hornbill says, “boom boom, I’m going home; boom, I’m going home” and the male counters with, “you always say that; boom, you always say that; I’m tired of hearing it; go on home; boom boom.” It was, therefore, with considerable interest that I learned from Mr. Rudyerd Boulton that the natives in Angola have another interpretation of the calling of these birds. They say that the female says, “boom boom, I’m going home, I’m going home,’ while the male replies with, “you must not do that, you must hold up the corn.” Like all hornbills these birds feed by picking up bits of food with the bill, then tossing it in the air and catching it far down in the bill or even in the open mouth as it descends. DISCUSSION OF HORNBILL BIOLOGY An early impression at Entebbe was that many of the nonmigratory tropical birds, from hadadas (Hagedashia hagedash) to red-bellied shrikes (Laniarius erythrogaster), remained paired throughout the year. Casqued hornbills were usually encountered in pairs. They are presumably mated for life and one would like to know when pair NO. 9 CASQUED HORNBILLS—KILHAM 4I formation takes place. It may have no immediate relation to the breed- ing season. I had three young captives, hand-reared and approxi- mately of the same age, in a cage at Entebbe. Mpanga and Zika were definitely paired before they were 3 months of age. Zika would work through Mpanga’s throat feathers as he let his head fall back, then he would do the same for her. The other hornbill, and later a fourth, both males, led independent lives except for roosting. Unnatural conditions undoubtedly favored this early pairing. Young birds, however, do not necessarily pair up when confined together. This was shown by three magpies (Pica pica hudsonia), taken before they left the nest, which I kept in a large cage in the same manner as the horn- bills. They showed no inclination to pair. Maintenance of close pair formation demands mutual attentions. When casqued hornbills are perching in different trees, the members of a pair are almost always in communication with each other, some- times only with single notes such as “‘cak” or “ugh.” When together, mutual preening, in which the female may take the lead, is a common activity. This preening about the head and nibbling of feathers under the throat went on regardless of the time of year. I saw it going on at dusk in the pair which roosted in our garden in October and again with the pair in the Botanical Gardens, on the day the female emerged with her young one in March. It took place early in the life of Mpanga and Zika. I have interpreted as courtship, activities which bring a pair of hornbills into the rhythm needed for the close cooperation involved in nesting. The lead is taken by the male. He feeds his mate and presents her with sticks and pieces of bark. In addition he becomes noisier in his calls and wailing. Similar activities are not uncommon to the courtship of many groups of birds. The male hornbill also takes the lead in exploring possible nest holes. By his cries and wailings, and his flights back and forth, he tries to induce his mate to look at them. Stonor (1937) has given an interesting account of a pair of trumpeter hornbills (Bycanistes buccinator) which attempted to breed in the London Zoological Gardens. Courtship consisted principally of the male feeding the female. She would fly down to the feeding dish, then wait expectantly for him to feed her. Sometimes he would do so. At other times he would swallow the food himself. Then, as if stricken with remorse, when she would fly to a higher perch, he would at once follow to feed her a morsel. Stonor wrote of a “curious ceremony, wherein the female flew up from the ground with food in her beak which she passed to the male, who then returned it to her, 42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 when she swallowed it.” I have previously described an almost identi- cal situation which took place in Mpanga Forest. Important differences in appearance of hornbills are located about the head, the region which can be seen best through a nest opening. Head feathers in birds of all ages express emotions. In young birds the feathers at the base of the upper mandible are brown instead of black. The huge, forward-projecting casque of the male is his chief sexual characteristic and white skin around the eye is a peculiarity of the female. Many African hornbills have brightly colored patches of skin and wattles about the head and neck. These, however, are en- tirely lacking in Bycanistes subcylindricus. One would like to know what part these bright colors may play in courtship performances. Coition in one pair of casqued hornbills took place without any special courtship other than some touching of bills. The pair were returning to their nest, after gathering termite earth. Moreau (1936) found that copulation took place in Bycanistes brevis just after the female had emerged from her morning’s work and about to days before the nest wall was complete. Good nesting sites are probably used annually. Pitman (personal communication, 1955) believed that the nest hole that I watched in the Botanical Gardens had been used in 1947 and in 1949. At nest 1 the pair tried for weeks to close the opening. Interest, however, began to fall off a week after coition, a situation that paralleled one de- scribed by Moreau (1936) in Usambara. Pairs of Bycanistes brevis tried for 2 years to nest at one site without success. “In both years,” Moreau wrote, “building continued after copulation had taken place, and when work had ceased, both birds still showed a keen interest in the nest hole.” Failure at the nest in Mpanga Forest may have been due to the large size of the opening. Other factors could have been operative also. The pair, or perhaps only the female, for example, may have been young and inexperienced. It is difficult to follow Moreau’s hypothesis that in Usambara, failure to complete nests was due to the male’s running out of saliva. Casqued hornbills probably lay two eggs to insure that a single healthy chick is produced. The young bird becomes so large that the nest might be overcrowded if two chicks survived. Crowned hornbills have two to three young. The mother, however, leaves the nest some weeks ahead of time. This not only makes more room for the young but enables her to help in the feeding. The length of time a female is walled in a nest (119 days for nest 5) does not appear unusual for a bird of hornbill size to lay eggs, incubate, and rear a young one. One can use Wahlberg’s eagle NO. 9 CASQUED HORNBILLS—KILHAM 43 (Aquila wahlbergi) for comparison. It is approximately the size of a casqued hornbill and has been studied by Leslie H. Brown (1952) in Kenya Colony. He observed an incubation period of 46 days and a fledgling period of 72 days at a nest where a single eaglet was raised. This gave a total of 118 days. The total time is about the same as for the casqued hornbill, which, I have estimated, leaves the nest when 10 to 11 weeks of age. I would have supposed that young hornbills would grow more slowly on a fruit diet—they had animal food only occasionally. However, as indicated by white matter in the feces, they may have had more animal protein, particularly in the form of smaller insects, than I realized. It was almost impossible to feed my young captive hornbills on fruit alone. The volume required was exhausting. We reduced the number of feedings, first by coating pieces of pawpaw with powdered milk, then by giving each bird six to eight half-grown mice a day. Intrusions of foreign hornbills on nesting pairs of their own species presented an interesting study. In a few instances the intruders came in pairs. There were many free pairs of hornbills throughout the nesting season and I wondered if these were not an index of an in- creasing shortage of suitable nest trees. Intrusions by single female hornbills were more difficult to explain. At one nest the same female apparently stayed around for months, and possibly attacked and crippled the young one soon after it left the nest. Several explana- tions suggest themselves. First, intruding females may have been offspring of the season before, unwilling to leave their parents or, second, they may have been unmated adults attracted by a seemingly lone male; possibly they fell into both categories. Some of them seemed to be more attracted to the nest itself and others to the male, coming and going with him as he made his feeding visits. I had an impression that there was an excess of females in the hornbill population. In conclusion, the pleasure of watching hornbills comes from their love of play, unexpected agility, clownishness, and seeming intelli- gence. This last quality is difficult to assess. The intelligence of the crow family is well recognized. I have kept tame blue jays, crows, and magpies and rate my captive hornbills on the same level. Both groups are playful and curious, examining new objects with interest. They have a wide range of vocal expression. The large eyes of horn- bills, together with expressive movements of head feathers, give an impression of intelligence which is hard to disregard. Whatever their mental capacity, however, it is difficult to see how these birds can adapt themselves to civilization, as they are destined to inhabit large 44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 forest trees. My unusual opportunity to study their breeding and other habits in the vicinity of Entebbe has been most fortunate. SUMMARY Sixteen nesting pairs of casqued hornbills (Bycanistes subcylin- dricus) have been studied in the vicinity of Entebbe, Uganda. An unusual concentration of nests was found in the Mpanga Research Forest. In courtship the male presented his mate with food and bits of bark or sticks. Mutual preening and calls back and forth were im- portant in maintenance of the pair bond. The male did the pioneering in exploration of possible nest sites and tried to entice the female to them with wailing cries. Plastering was done by the female from the inside of the nest cavity, using a rapid tapping with the side of her bill tip. Both sexes flew to the ground to collect dirt for building. Most of this, however, was done by the male. He attended the female, furnishing her with pellets for construction of the nest wall. Termite earth was preferred for building. Two eggs are laid. Although both may hatch, only one chick is raised. Male hornbills feed their mates about every 30 to 60 minutes, heaving up fruits held in their gullets. Often a stick or piece of bark is presented first. Nest sanitation is accomplished by female and chick expelling feces through the entrance. The female may also toss debris out with her bill. If such debris collects outside of the entrance, it is regularly cleared away by the male. Ants which swarm in some hornbill nests may act as scavengers and keep down insect fauna. Females walled within nests can be extremely noisy if alarmed. Observations made at one-third and two-thirds through the nesting period indicate that females have a gradual molt. Two females re- moved when two-thirds through their nesting could fly readily. The total period a female was walled in one nest was I119+2 days. A young hornbill, captured two days after emerging, was in full plumage. The majority of casqued hornbills in eastern Uganda probably begin nesting in September and break out in January. A hornbill territory is confined largely to the vicinity of the nest tree. NO. 9 CASQUED HORNBILLS—KILHAM 45 Occasionally a foreign pair of Bycanistes would visit a nesting female and attempt to feed her. Lone female hornbills interfered persistently with a number of nesting pairs. At one nest this interference had serious consequences. Male hornbills fed their mates largely on fruit but also caught in- sect and vertebrate prey. Lists are given of such fruits and insect remains as could be identified. Roosting habits are described for one pair before and for a lone male during the nesting season. Casqued hornbills were much alarmed by hawks and eagles. They prefer the largest of forest trees to nest in. Rapid destruction of forests in Uganda threatens the future of these birds. Three hornbills, removed from nests when 6 to 7 weeks of age, have been reared in captivity, largely on a diet of animal protein. The exact age of one captive was known. Two of them have remained closely paired from the age of 3 months. Bill tapping and plastering against the walls of their cage were done by the male and the female, beginning at 9 to 10 months of age. A comparison has been made of Bycanistes and Tockus. There are many points of similarity in the habits of the two genera. REFERENCES BANNERMAN, D. A. 1953. The birds of West and Equatorial Africa. 2 vols., 1,526 pp., 144 figs., 54 pls. Brown, Les.ie H. 1952. On the biology of the large birds of prey of the Embu district, Kenya Colony. Ibis, vol. 94, pp. 577-620. MackwortH-Praep, Cyrit W., and Grant, C. H. B. 1952. Birds of eastern and northeastern Africa. xxv + 836 pp., 53 col. pls., 6 black-and-white pls., hundreds of text figs. and maps. Moreau, R. E. 1936. The breeding biology of certain East African hornbills (Bucerotidae). Journ. East African and Uganda Nat. Hist. Soc., vol. 13, Nos. 1 and 2, pp. 1-28. RANGER, G. 1949a. Life of the crowned hornbill, Lophoceros suahelicus australis. Terri- torialism, family life, and breeding. Ostrich, vol. 10, pp. 54-65. 1949b. Life of the crowned hornbill (Pt. II). Ostrich, vol. 20, pp. 152-167. 1950. Life of the crowned hornbill (Pt. III). Ostrich, vol. 21, pp. 2-14. 1951. Life of the crowned hornbill (Pt. IV). Ostrich, vol. 22, pp. 77-93. 1952. Life of the crowned hornbill (Pt. V). Ostrich, vol. 23, pp. 26-36. Stonor, C. R. 1937. On the attempted breeding of a pair of trumpeter hornbills (By- canistes buccinator) in the Gardens in 1936, together with some remarks on the physiology of the moult in the female. Proc. Zool. Soc. London, vol. 107A, pp. 89-95. a on ae, Pine ot 4 mt 7 * i i Hy ae i J : B é i? ' / 7 7 * , * rp ey . of yet ot rit , ' oo. . Seren : “ rat cae = . | ihe ct ch: m wi. 4 aad f ered ps Z : _ 7. pa ; ud i a . 7 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 10 CRUSTACEAN METAMORPHOSES By R. E. SNODGRASS Collaborator of the Smithsonian Institution and of the U.S. Department of Agriculture (PusiicaTion 4260) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION OCTOBER 17, 1956 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U. S. A. CONTENTS MER OCUCEION: o's .c\a/s aye, sieve store ssa icicicl alerts iebe ltrs ars sues ie Evolution o1 the*arthropodS2:/s.e. tees. ciere'e cies II. The nauplius and the metanauplius............. III. Examples of crustacean metamorphoses......... Bravichtopodal << jac,cceyesosia sehecise/etere estes eae Ostracoda. swe sslowis she ccusa leis aoe ise se Se CODED OA ichaevesteciss dt aeiorcioste amines oi aw oe CirKipediay Se Ps we 5 homer nears seraiorearetes PSOPOGAs Micra seteleiciarsiotererseieto seajetel a sie sia thale. ster IF tgp paAiSta CCD dive laraaroneyalereresval orate tel avovcuele- serie Wecanoda: yee.s ais teins wie oie Sera cisidislow os. ttoretelorers StOMALOPOGA: | sie cievsist cverusnasis dele iacrwievs aula! sii IV. Structure and evolution of arthropod appendages RELEKENCES). te cistarccoets cist oahetereis nial cletole oh werk stores © 210,6/0).8) 68. @; 010'(8, 68 «09.0 see ee eee eee eee eee oe ee a), 06. 82, 0 ahe eee) see ai eis\s 6s) \e 0) 6 \6\0) 6, 66 [6)'e" 6 eet eee eee eee eet one 06 4, eee a) 6 66 @, 60a 15,.a ele. eee ee wee te eee ee eee Se ee ewe ee eee eee tee eee ee wee eee eee ee ee eee ewe Oo a 0.67 8:0 16,\6 0) 60 0 0,0 6 6's €)10)"6i @ (01.6 16:00, 6/6 6: 6 8 © a 0.8 0100: 10.0.6) 016 (60,0, B.6 ce @ ® eee eee errr ee eeetnne iii » i 4 Ay APT _ a ‘ teivhe . Won , ae iil hé ith 7 0 yg ie ‘| ¥ i” y lo Gi " * an ’ i: ad A edit: vn is Wipiee & peat a ed ' of, , ae nat Nip a ee | 7 _ - » Py | ro ' # AL yoy tC , i i ahst* 7 ’ a : Pe i. ‘ pare # (ow tts k, : poy anne i i) o. / - aa ; 7 i | 7 : vue ¥ : a | _ - 4 - : CRUSTACEAN METAMORPHOSES By R. E, SNODGRASS Collaborator of the Smithsonian Institution and of the U.S. Department of Agriculture INTRODUCTION The review of crustacean metamorphoses given in this paper con- tains little that will be new to carcinologists, except perhaps a few accompanying unorthodox ideas. The paper is written for students in general zoology and is recommended reading for entomologists, who commonly think of metamorphosis as a phenomenon pertaining par- ticularly to insects. It is true that the metamorphoses of insects and of crustaceans have no relation to each other, and have little in com- mon, but a preliminary discussion of both will help in arriving at a general understanding of the nature of metamorphosis as it occurs in the arthropods. The first treatise on metamorphosis was written by Ovid in about the year A.D. 7, but the metamorphoses that Ovid described were mostly the transformations of members of the human species into animals, trees, or rocks, willed by the ancient gods or goddesses in revenge against some mortal that had offended them. The meta- morphoses imposed on animals by nature are not punishments, ex- treme as they may be in some cases, but are beneficent changes of form to better accommodate the individuals of a species temporarily to a more advantageous way of living. The young butterfly, for example, transformed in the egg into a wormlike caterpillar, is not an elegant creature as are its parents, but from a practical standpoint the cater- pillar is perfectly adapted to its chief function, which is that of feeding. The metamorphoses of Crustacea differ essentially from those of insects in that they pertain to a much earlier stage of development. The young insect hatches from the egg usually with the definitive number of body segments. The insects are thus epimorphic; but if the young insect has taken on a metamorphosed form in its embryonic development, it appears on hatching as a creature quite different from its parents. Yet a caterpillar, for example, is actually a winged juvenile stage of the butterfly corresponding with the so-called nymphal stage SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 131, NO. 10 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I of a grasshopper. The principal difference between the two is that the wings of the young grasshopper develop externally, and those of the caterpillar grow internally within pockets of the skin beneath the cuticle. Likewise, a “legless” fly maggot has legs developing in pouches of the skin covered by the cuticle. A young grasshopper goes over directly into a mature grasshopper; the caterpillar, the maggot, and others of their kind, when full grown with plenty of food stored in their bodies, must undergo a second transformation in a pupal stage to be restored to the parental form. This is the usual course of metamorphosis among the insects. Most of the Crustacea, on the other hand, hatch at an early stage of embryonic development, though at varying periods of immaturity, when they have only a few body segments and corresponding ap- pendages. During their development after hatching they successively add new segments and appendages until the definitive number is attained. The majority of crustaceans are thus anamorphic in their manner of postembryonic growth, though a few are epimorphic. Anamorphosis involves a change of form during development, but it is merely a way of growing, common to crustaceans, diplopods, and some chilopods. It should not be confused with changes of form that have nothing to do with progressive development toward the adult; such changes constitute a true metamorphosis. The metamorphoses of Crustacea are changes of form that the growing animal may take on at successive stages of its anamorphic growth, including the sex- ually mature stage of many parasitic species. In such cases, meta- morphosis has been superposed on anamorphosis. As Gurney (1942) has said, “it may be assumed that development in the Crustacea was primitively a continuous process of growth and addition of somites and limbs, as we find it to be in some branchiopods, and that abrupt changes between successive moults leading to the origin of definable phases are secondary responses to changes in the habit of life of the larva and adult.” Gurney notes an apparent exception to this rule in the Euphausiacea and some Penaeidae, in which the larva and the adult lead much the same kind of life. The successive phases of de- velopment in these two groups, however, are mainly stages of ana- morphic growth; their only metamorphosis is the adaptation of the larval appendages for swimming. Insect larvae may undergo metamorphic changes of form during their growth, but with the insects this larval heteromorphosis, com- monly called “hypermetamorphosis,” affects the fully segmented young insect, and is therefore not comparable to the heteromorphic larval growth of most Crustacea. Some metamorphosed young insects trans- NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 3 form directly into the adult, but most of them first undergo a recon- struction in a special, proimaginal pupal stage. Among the Crustacea there is no transformation stage strictly comparable to the insect pupa. True metamorphic forms are not recapitulations of phylogenetic stages in the evolution of a species. An insect larva, though often . wormlike in appearance, does not represent a worm stage in the an- cestry of insects. A caterpillar has a modern insect head and mouth parts, a well-developed tracheal system, and wings growing beneath its cuticle. No worm, ancestral or otherwise, was ever thus equipped. Among the Crustacea also most juvenile forms assumed during the larval growth are temporary adaptations to a changed mode of life and are not phylogenetic recapitulations. Yet, it is true that former ancestral characters discarded somewhere along the line of evolution may appear in the ontogeny of the individual, and it is often difficult to determine what phases of development are recapitulatory and what are metamorphic aberrations. The following hypothetical example will make clear the distinction between the two, and will lead to a practical definition of metamorphosis. If the eggs of birds regularly hatched into reptilelike creatures, which later transformed into feathered fowls, the change of form would literally be a metamorphosis ; but, since birds have been derived from reptilian ancestors, it might be specifically a case of phylogenetic recapitulation. On the other hand, if there issued from the bird’s egg a creature having no relation to anything in the avian line of adult evolution, but which still finally transformed into a bird, the change of form would be one of quite a different nature, and it is this kind of change that will be regarded as metamorphic in the following dis- cussions. As here defined, therefore, metamorphosis is a structural change at any time in the life history of an animal that may be re- garded as an aberration from the ancestral direct line of adult develop- ment which followed approximately the phylogenetic course of evolu- tion of the species. In this case metamorphosis may affect the embryo, the larva, or the adult. Simple development without metamorphic interpolations might then be termed orthomorphosts. In the higher Crustacea there is a tendency for hatching to take place at later and later stages of ontogeny, leaving a correspondingly lesser amount of development to be accomplished after the larva leaves the egg. Finally a condition is reached when body segmentation and ap- pendage formation are complete or almost so at hatching ; the animal then becomes epimorphic in its development. In an epimorphic arthro- pod, the embryonic development may proceed by the method of ana- morphosis, or the entire body may be first laid down as a germ band. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I In the second case segmentation appears later, usually progressing from before backward, suggesting that it represents a former ana- morphic mode of segment formation in which the anterior segments are the oldest. Since anamorphic growth, either in the larva or the embryo, is characteristic of the annelid worms and recurs in so many of the arthropods, it was probably the primitive method of growth in the annulate animals. The most immature larval form among the arthropods is the crus- tacean nauplius. For practicable purposes early hatching must be given up by terrestrial animals, unless they go back to the water to lay their eggs, as do the land crabs, frogs, and toads. The anamorphic myriapods do not quit the egg until they have acquired the adult type of structure and are equipped with a sufficient number of legs for terrestrial locomotion. The completely epimorphic spiders and insects are best fitted to cope at once on hatching with the conditions of their environment, and they have become the most successful of the land arthropods. Though some insects lay their eggs in the water and the young are aquatic, they are simply terrestrial forms that have become secondarily adapted in the larval stage for life in the water; they hatch at the same stage as their relatives on land. The Crustacea are primarily aquatic animals; only a few have be- come adapted to a permanent life on land. The eggs of most species are laid in the water, and the newly hatched young must be capable of swimming; the adults can later adopt a bottom habitat if they ac- quire ambulatory legs. Considering the uniformity of the water environment of a swimming larva, there is relatively little inducement for a young aquatic animal to undergo adaptive metamorphoses. The metamorphoses of most crustacean larvae, therefore, are relatively simple as compared with those of insect larvae, which have a great diversity of habitats open to them. Parasitic crustaceans, however, are a conspicuous exception to this generalization. As a rule small animals in the water are eaten by larger animals, but the small creatures have one recourse against their possible preda- tors and that is to become parasitic on them. Parasites, however, have to be structurally adapted to a parasitic life, and consequently most parasites undergo metamorphic changes. Many of the smaller crustaceans have adopted parasitism, and the most extreme degrees of crustacean metamorphosis are found among such species, especially if the adults themselves remain parasites. Such adults in some cases have lost all resemblance to the ancestral forms of their race, even every mark of their crustacean ancestry. Moralists may cite the “degeneration” of such parasites as a warning of what parasitism may NOs /LO CRUSTACEAN METAMORPHOSES—SNODGRASS 5 lead to, but actually parasites are highly specialized for the life they lead by a simplification of structure and the elimination of all un- necessary organs, which were indispensable to their free-living an- cestors. In fact, no sympathy need be wasted on “degenerate’’ para- sites; give them credit for having found a simple and easy way of living, though at the expense of another creature. They have discarded all useless equipment, and some of them have devised most ingenious ways of attacking the host. The control of metamorphosis by hormones has been extensively studied in insects, but apparently no comparable studies have been made on the role of hormones in the metamorphosis of crustaceans. It is well known that hormones are produced in the eyestalks of decapods, and the source of the eyestalk hormones has usually been referred to two organs known as the sinus gland and the X organ. However, from recent investigations (see Bliss and Welsh, 1952; Passano, 1953) it is now known that the so-called sinus “gland” is not a gland but a complex of the enlarged ends of nerve fibers pro- ceeding from the X organ and from numerous neurosecretory cells in the brain, in the ganglia of the optic lobe, and in the thoracic ganglia. The sinus “gland” is therefore a receiving and distributing center for various hormones. Functions that have been attributed to these hormones include the movement of pigment in the compound eye, regulation of chromatophore activity in the integument, control of moulting, and the rate of development of the ovaries. Knowles (1953) gives evidence that the chromatophores are activated also by neurosecretory cells in the region of the tritocerebral commissure and the postcommissural nerves. The control of moulting by lengthening the period between moults was attributed by Passano to the X organ, which is itself a neurosecretory tissue in the proximal ventral part of the medulla terminalis of the optic lobe. Removal of both sinus “glands” has no effect on moulting since the hormone can escape from the cut ends of the nerves. Panouse (1946) also, in a study of Leander, had claimed that the “sinus gland” produces a hormone that normally blocks the growth of tissues and thus causes a lengthening of the intermoult period and retards the maturing of the ovaries. From later work by Gabe (1953) and Echalier (1954), however, it now appears that moulting, at least in the Malacostraca, is controlled by a pair of ductless glands in the antenno-maxillary region. These glands, discovered by Gabe, are named by him the Y organs, and were demonstrated to be present in 58 malacostracan species, ranging from Nebalia to the decapods and stomatopods. In species in which the excretory gland is maxillary, the Y organs are in the antennal seg- 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 ment ; in those having antennal glands they lie in the second maxillary segment. Each gland is implanted on the epidermis by an enlarged base and is innervated from the suboesophageal ganglion ; in form it is conical, lenticular, or foliaceous. From their histological structure and changes during the intermoult period, Gabe shows that the Y organs are comparable to the thoracic endocrine glands of holometabolous insects, and he suggests that they have something to do with moulting. Following this suggestion, Echalier (1954) made experimental tests by removing the organs. He found that bilateral ablation of the glands, when not made too late after they had already discharged their secretion, resulted in a very great lengthening of the intermoult period, far in excess of the usual time between moults. Echalier, therefore, contends that the Y organs are crustacean endocrine glands for the control of moulting. That they do not disappear in the adult as do the thoracic glands of insects, Gabe points out, follows from the fact that the crustaceans continue to moult in the adult stage. I. EVOLUTION OF THE ARTHROPODS In any discussion of arthropod metamorphosis the question of re- capitulation always comes up in relation to the larval forms. If there is any ancestral recapitulation in ontogeny, it then becomes necessary to have at least a theoretical concept of the evolution of the arthropods and some idea of what ancestral forms they had that might be re- capitulated in the development of the individual. The evolutionary origin of the arthropods is hidden in remote Pre- Cambrian times, so probably we shall never know the facts from visual evidence. There is, however, ample evidence from a study of modern forms to indicate that the early progenitors of the arthropods were closely related to the progenitors of the annelid worms, and that these two groups of annulate animals had a common ancestor. The funda- mental characters preserved in the annelid—arthropod organization are: an elongate segmented body, an alimentary canal extending through the length of the body, a paired ventral nerve cord with seg- mental ganglia, a somatic musculature, and mesodermal coelomic sacs. We may therefore visualize the primitive annulate as a very simple, wormlike creature having these features. The mode of development was anamorphic, new segments being formed in a subterminal zone of growth. From this primitive segmented worm the annelids have been directly evolved with little addition other than the development of segmental groups of lateral bristles, which in the polychaetes have been carried outon movable lateral lobes of the segments, the so-called parapodia, that serve for swimming and burrowing. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 7. By a different type of specialization for locomotion, members of another branch from the ancestral stock developed ventrolateral, lobelike outgrowths of the body segments, and thus became walking animals. These primitive legs eventually evolved into the jointed appendages of modern arthropods, the lobelike origin of which is still recapitulated in the embryo. At the lobopod stage of evolution (fig.1 A) the animals resembled a modern onychophoran, and are Fic. 1.—Theoretical evolutionary stages of the arthropods. A, a primitive lobopod, common ancestral form of the Onychophora and Arthropoda. B, a derived form with longer and slenderer legs. C, a primitive arthropod with sclerotized integument, jointed legs, and gill lobes on the coxae. D, a fairly generalized modern crustacean, Anaspides tasmaniae. rAnt, first antenna; 2A4nt, second antenna; Mxpd, maxilliped; Plpds, pleo- pods; Prpds, pereiopods; Tel, telson; Urpd, uropod; IJ-XV III, body segments. perhaps represented by such fossils as the Pre-Cambrian Xenusion and the Cambrian Aysheaia. The modern Onychophora are probably direct descendants from these early lobopods, and have structurally not progressed much beyond them. Others, however, acquired a sclerotization of the integument, which allowed the legs to become longer and slenderer (B), and finally jointed (C) for more efficient action in locomotion. These jointed-legged forms were the first true arthropods. The segmentation of the legs early took on a definite pattern, which has been preserved in both fossil and living arthropods, most of which retained the walking mode of locomotion, though some may also swim or fly. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I From these early Pre-Cambrian arthropods (fig. 1 C) in which all the appendages were fully segmented ambulatory legs, the trilobites branched off by specialization of the body structure, but with no es- sential differentiation of the appendages. In the other derivative groups, however, the appendages took on different forms adapting them to various uses, but the number retained for walking is charac- teristic of the several modern arthropod groups. The myriapods use most of their postoral appendages for progression ; the Malacostraca (D) use five or more pairs for walking, except where some of these have been modified for grasping ; Limulus and the arachnids use four pairs, the insects three. That the ambulatory limbs, when limited in number, should in all cases be those of the middle part of the body, though not necessarily the same appendages, ‘follows from the me- chanical necessity of balance. The anterior appendages become sen- sory and gnathal in function; those of the abdomen have been modi- fied for various purposes, such as respiration, silk spinning, copulation, egg laying, or swimming. The modern arthropods comprise two distinct groups, the Chelicer- ata and the Mandibulata. In the chelicerates the first postoral ap- pendages are a pair of pincerlike chelicerae that serve for feeding, and the ancestors of this group were probably closely related to the ancestors of the trilobites. The principal feeding organs of the mandibulates are a pair of jaws, the mandibles, formed of the second postoral appendages. The Mandibulata, including the crustaceans, the myriapods, and the insects, are certainly a monophyletic group, but their origin and their interrelationships are obscure. Among the Crustacea the malacostracan type of organization (fig. 1 D), in which the thoracic appendages are typically ambulatory and the abdominal appendages natatory, would appear to be more primitive than the entomostracan types because it more closely conforms with the structure of other arthropods, and could be more directly derived from that of a primitive walking arthropod (C). The entomostracan forms, therefore, have been secondarily reconstructed for a purely pelagic life by a readaptation of the thoracic appendages for swimming. If we accept the premise that the original arthropod (fig. 1 C) was a simple animal with jointed legs along the entire length of a uniformly segmented body, the crustaceans were derived from this common arthropod ancestor by specializations that established the generalized crustacean structure (D). Developmental recapitulation of adult crustacean structures, therefore, can go back only to the beginning of adult crustacean evolution. The embryo, however, starts its de- velopment from a single cell and the free larva completes development NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 9 up to the adult. The embryo and the early larva, therefore, represent pre-crustacean stages of arthropod evolution. The embryo, however, must reproduce its parental form. Hence the crustacean characters appear at a very early stage of ontogeny, but the resulting embryonic or larval stages are not recapitulations of adult crustacean evolution. The crustacean characters are simply precociously imposed on the anamorphic stages of ontogeny. Finally, if the embryo is set free as a larva at an early stage of development, it must be structurally adapted to a free life, and in its subsequent growth other adaptations may be necessary. Thus it comes about that metamorphosis still further complicates the course of ontogeny. The life histories of parasitic larvae best demonstrate that larval forms are metamorphic adaptations to a way of living, since the nonparasitic adult ancestors of such spe- cies can hardly be supposed to have had the larval form. Where a specialized adult structure has arisen since the crustaceans became crustaceans, there may be a true recapitulation of an earlier adult form, as in the megalops of the crabs. A further discussion of the nature of larval forms will be given in connection with the life history of a penaeid (p. 54). II. THE NAUPLIUS AND THE METANAUPLIUS Since among the crustaceans the young hatch at different periods of development, the youngest larvae may have very diverse forms in the various orders, representing different ontogenetic stages according to the degree of development they undergo within the egg. The earliest hatched larval form is the nauplius, which is particularly characteristic of the Entomostraca, but occurs also in the Euphausiacea and Pe- naeidea among the Malacostraca. The nauplius is usually followed by a metanauplius, which is the first stage of postembryonic growth. From the metanauplius on, development may be merely a matter of regular anamorphic growth by the successive addition of new segments and appendages, but in many species the larva takes on different forms as it develops. These ontogenetic changes differ so much in the various orders that no general description can be given, hence a discussion of them will be left to the next section of this paper. Special attention, however, must be given to the nauplius and the metanauplius. The nauplius—The nauplius is a minute creature, highly variable in form in different species, but typically ovoid or pyriform in shape with the larger end anterior (fig. 2 A). It has a pair of uniramous an- tennules, or first antenna (1Ant), typically biramous second antennae (2Ant) and mandibles (/d), and a median eye of two or more parts. Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31T The antennae serve for locomotion. The internal organization includes usually an alimentary canal, a muscular system, a nervous system, and a pair of antennal excretory glands. The alimentary canal when fully developed consists of an endodermal mesenteron and an ectodermal stomodaeum and proctodaeum. The mouth is concealed above a large labrum; the anus is usually formed at a later stage. The nervous system includes three pairs of ganglia corresponding to the append- ages. Though there is no visible segmentation in the ectoderm, the presence of appendages and ganglia shows that the nauplius is at Fic. 2—Nauplius and metanauplius of Apus cancriformis Bosc (from Claus, 1873). A, nauplius, with first antennae (7Ant), second antennae (2Ant), and man- dibles (Md), rudiments of teloblastic appendages seen through cuticle. B, metanauplius, appendages of teloblastic segments (thSegs) exposed after first moult. least a partly segmented stage of development. The region of the body behind the mandibles is that in which later the other segments will be formed, and their rudiments may be seen beneath the naupliar cuticle. When these segments are formed, however, they are generated by a different method from that which formed the anterior segments. The nauplius is derived from a very early stage of embryonic de- velopment, represented in species that hatch at a later period by a simple embryo with rudiments of three pairs of appendages. The embryo still in the egg at this stage is clearly more simple in its structure than is the nauplius. The nauplius, therefore, is not merely an early hatched embryo—it has undergone a metamorphosis before hatching to adapt it to a free life in the water. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS Il Our chief interest in the nauplius is the question of its theoretical value in phylogeny. The nauplius has been likened to the polychaete trochophore, and has been regarded as representing a primitive an- cestral form of the Crustacea. The trochophore, however, is entirely unsegmented and does not have the internal organization of the nauplius. Later it becomes segmented by a direct division of the posterior part of its body into a few primary somites. Likewise the very young trilobite, known as a protaspis, at first shows no sign of segmentation, but it soon becomes marked by transverse grooves that divide it into a few primary segments corresponding with the segments in the prosoma of the adult. A similar early direct segmenta- tion occurs also in the ontogeny of the Xiphosurida. The nauplius, therefore, would appear to represent the same stage of primary seg- mentation in crustacean ontogeny, though metamerism has not yet affected the ectoderm. It is reasonable then to infer, as contended by Iwanoff (1928), that the first somites in both the annelids and the arthropods were formed directly in the previously unsegmented body of the animal. The later extension of the body took place by the teloblastic generation of secondary somites from a subterminal zone of growth. The annelid and arthropod ancestors did not diverge until this method of anamorphic growth was fully established. While the three larval forms discussed above do have a basic simi- larity of structure, which is primitive, it is evident that distinctive characters of more recent phylogenetic evolution have been impressed separately on each. The protaspis shows distinctly the definitive trilobite type of structure, the nauplius is clearly a crustacean, the trochophore is a young worm. The trochophore and the nauplius, moreover, are adapted in quite different ways for swimming at an early ontogenetic stage. The trochophore is not an adult ancestral form of the annelids, nor is the nauplius an ancestral form of the Crustacea. The metanauplius—The nauplius is the direct product of em- bryonic development. The further growth of the larva, or of the embryo if hatching occurs at a later stage, proceeds from a subterminal zone of growth, which becomes active before the naupliar cuticle is shed, so that rudiments of the new segments may be seen in the posterior part of the body of the nauplius (fig. 2A). In the meta- nauplius (B), which appears after the last ecdysis of the nauplius, the posterior part of the body is much lengthened; it is now distinctly segmented and bears the rudiments of several pairs of new append- ages. The postmandibular somites are the teloblastic segments (tbhSegs). I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 Most studies that have been made on the larval development of Crustacea describe and picture the newly forming postnaupliar somites and appendages as they appear externally, without giving any in- formation as to how they are formed. A paper by Fransemeier (1939), however, describes the segment formation in the free-swim- ming metanauplius of the branchiopod Artemia salina, and papers by Sollaud (1923) and by Manton (1928, 1934) give details of the corresponding segmentation in the embryos of Leander, Hemimysis, and Nebalia. In the anterior part of the nauplius the embryonic ectoderm is already differentiated into the tissues derived from it, and the meso- derm has been formed from the embryonic mesoblasts. The ectoderm of the body region behind the mandibles, however, is still undiffer- entiated and there is here no mesoderm distinguishable at this stage. At the posterior end of the body of Artemia the ectoderm forms a circumanal fold, the cells of which are the ectodermal teloblasts that will form the ectoderm of the new segments. From the ectodermal teloblasts, according to Fransemeier, cells are given off into the in- terior of the body that become the mesodermal teloblasts, which will generate the secondary mesoderm. The naupliar mesoderm and the postnaupliar mesoderm of Artemia are thus distinct in their origin, though the formation of the second takes place 10 to 15 hours before the hatching of the nauplius. The teloblasts constitute the zone of growth, from which the new segments will be generated forward. The first segments formed from the teloblasts are said by Fransemeier to be those of the first and second maxillae. As other segments are generated the anus-bearing region is carried posteriorly as a permanent telson. The proliferation zone remains active until the last segment is formed, when it is fully exhausted. The alimentary canal apparently simply lengthens posteriorly, the proctodaeum having been formed in the nauplius. In the young naupliar embryo of the palemonine Leander, as de- scribed by Sollaud (1923), the postmandibular part of the body is a small anus-bearing lobe, or caudal papilla, which subsequently lengthens and projects free from the body in front of it and bends forward. A transverse row of large cells becomes differentiated in the ectoderm of the lobe before the anus, and later encircles the lobe. These cells are the ectodermal teloblasts. Below and a little before them is formed a corresponding ring of mesodermal teloblasts, which, according to Sollaud, are derived from the blastopore. The teloblasts generate the secondary segments in the usual manner, but in Leander, Sollaud says, the two maxillary segments are formed directly in the NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 13 larval body and not from the teloblasts, the first teloblastic segment being that of the first maxillipeds. In the Pericarida, however, he says the boundary between the primary tissue and the secondary tissue is between the segment of the mandibles and that of the first maxillae. The accounts given by Manton (1928, 1934) of the embryonic process of secondary segmentation in Hemimysis and Nebalia are essentially the same as those of Sollaud for the embryo of Leander and of Fransemeier for the larva of Artemia, Manton agrees with Fransemeier that the teloblastic segments include both maxillary seg- ments. In Hemimysis, she says, the naupliar and postnaupliar meso- derms are at first some distance apart, but later the teloblastic ecto- derm and mesoderm extend forward as far as the first maxillary segment inclusive. The teloblasts of Nebalia are differentiated at the sides of the posterior blastoporic area, and the ectodermal teloblasts eventually form a complete circle around it. The mesodermal telo- blasts, according to Manton, in agreement with Sollaud, are formed from the mesendodermal mass at the blastopore; Fransemeier says they are proliferated from the ectodermal teloblasts. The ectodermal teloblasts, according to Manton, join the naupliar ectoderm between the mandibular and first maxillary segments, so that “all segments between the mandibular segment and the telson are formed by the teloblasts.” The rudiment of each segment arises from one transverse row of descendants from the original ectodermal and mesodermal teloblasts. When the last abdominal segment is completed the telo- blasts disappear in both Hemimysis and Nebalia. Since the teloblastic generation of secondary somites added to the primary segmented body of the young larva or embryo is characteris- tic of the annelid worms and recurs in many of the arthropods, it must have been a way of lengthening the body developed in the very primi- tive wormlike ancestors of the two groups. The annelids and the arthropods, therefore, did not diverge until this method of growth was well established. Elsewhere the writer (1938) has suggested that telogenesis may have originated as a means of increasing the repro- ductive function by distributing the germ cells from the zone of growth through a larger number of segments. III. EXAMPLES OF CRUSTACEAN METAMORPHOSES The metamorphoses of Crustacea are so diverse that in a brief re- view of the subject we can include only a few examples representative of some of the principal orders. Since crustaceans that hatch at an early stage of ontogeny go through anamorphic phases of development I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I by the successive addition of segments and appendages, many of their changes are merely those resulting from the anamorphic manner of growth. In nearly all cases, however, there is some degree of meta- morphosis superposed on the anamorphic stages, varying from a mere adaptation of the appendages for swimming to a total recon- struction of the animal for a parasitic way of life. The most striking examples of crustacean metamorphosis, therefore, occur in parasitic species. Among the Crustacea metamorphosis evidently has been de- veloped separately in each order, and often independently in different members of the same order. There is no type of metamorphosis characteristic of large groups of orders, as in the holometabolous orders of insects. Moreover, since crustacean metamorphosis affects the juvenile anamorphic stages, except where it is carried over into the adult, the metamorphosis of Crustacea has no relation to that of the epimorphic insects. A brief but interesting account of the life- history problems of crustacean larvae is given by Gurney (1926). For much assistance in preparing this section of the paper the writer is indebted to Dr. Fenner A. Chace, Jr., and his associates in the di- vision of marine invertebrates of the U. S. National Museum. BRANCHIOPODA The branchiopods undergo few changes during their larval de- velopment that are not the result of simple anamorphic growth by which the body and the appendages are completed and brought to the adult condition through successive instars. The branchiopods are thus of interest in showing a simple progressive development from nauplius to adult, which is accompanied, however, by a specialization of the postgnathal appendages for swimming. As an example we may take the life history of Branchinecta occidentalis Dodds as described by Heath (1924). The newly hatched larva of Branchinecta is a typical nauplius (fig. 3 A) with three pairs of appendages, a median simple eye, and a large labrum, but the oval, unsegmented posterior part of the body is more than usually constricted from the forepart. The large second antennae are the principal swimming organs. Between the nauplius and the second instar, or metanauplius (B), a very considerable change takes place. Lateral compound eyes are now conspicuous by their pigmentation. The posterior part of the body has greatly length- ened, and bears rudiments of maxillulae, maxillae, and six or seven following pairs of appendages. In the third instar (C) the post- maxillary appendages have lengthened and the more anterior pairs NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS 15 have taken on a leglike form; their mesal margins are indented, and each limb bears a conspicuous lobe, or flabellum, just proximal to an apical point. The body is more lengthened behind the appendages and Fic. 3.—Branchiopoda. Branchinecta occidentalis Dodds, developmental stages (from Heath, 1924) and a thoracic limb of the adult. A, newly hatched nauplius, length 0.4 mm. B, second instar. C, third instar. D, fifth instar. E, eighth instar, 29 mm. F, sixth left thoracic limb of adult male, with six endites (1-6) and a movable terminal lobe (Dactpd). G, head of adult male, anterior, with large second antennae (2Ant). shows lines of further segmentation. After two more moults, the larva in the fifth instar (D) reaches an average length of 1.6 mm. The second antennae are relatively much shortened, but the legs have 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I increased in length and are more distinctly indented on their mesal margins. The slender posterior part of the body bears rudiments of four new appendages, and its apex is split into a pair of small caudal lobes. The postmaxillary appendages continue to develop through the sixth and seventh instars until 11 pairs are present. In the eighth instar (E) they have lost their leglike form and have become broad flat phyllopodia with large flabella and slender apical lobes. At this stage, as the thoracic appendages take over the swimming function, the second antennae are much reduced in size and are directed for- ward. Heath enumerates 17 instars in the larval life of Branchinecta, but development beyond the eighth instar merely brings about refine- ments toward the adult structure. The larval stages of Artemia described by Heath (1924) are very similar to those of Branchinecta, as are those of Branchipus described by Oehmichen (1921). In the Concostraca and Cladocera the larval development is complicated by the formation of a bivalved shell. The development of the branchiopod appendages is of interest be- cause it suggests that the natatory phyllopodium has been evolved from a segmented ambulatory leg. The mature appendage of Branchi- necta (fig. 3 F) is cut on its mesal margin into a number of lobes, of which five (7-5) are commonly described as endites, while the large, so-called flabellum (6) is interpreted as the endopodite, and the mov- able apical lobe as the exopodite. The same structure is seen in the limbs of Branchipus (fig. 27 A,B) and other anostracans. Since endites in general are lobes of the limb segments, the six mesal lobes of the phyllopodium suggest that they represent six leg segments, coxopodite to propodite. The movable, independently musculated apical lobe (Dactpd), therefore, should be the dactylopodite. There is thus in the phyllopodium evidence of the presence of the seven segments characteristic of the crustacean walking legs. In the second maxilliped of Apus (fig. 27 C) seven segments, including a terminal dactylopodite, are plainly evident, and each of the first six segments except the ischiopodite bears an endite. We can hardly escape the conclusion, therefore, that the phyllopodial limbs of the branchiopods have been evolved from 7-segmented walking legs. The metamorpho- sis of the appendages, therefore, has taken place since the crustaceans became crustaceans, and is recapitulated in the larval ontogeny. A more extensive discussion of the nature of the primitive arthropod limbs is given in section IV of this paper. About the only metamorphosis in the life history of Branchinecta is the temporary adaptation of the antennae for swimming. It is hardly to be supposed that the primitive crustaceans swam with their NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 17 antennae. The nauplius has only three pairs of limbs, and, since it must swim, it has no choice but to use what appendages it has. As the body lengthens and the postmandibular appendages become broad and flat, these appendages assume the function for which they were modified in the branchiopod ancestors. The antennae then revert to a more simple form (fig. 3 E), and in the adult they are again modi- fied, in the male (G) for grasping. OSTRACODA The ostracods, being enclosed in a bivalve shell from the time they leave the egg, go through no body changes of form that might be termed a metamorphosis ; their appendages, however, give an impres- sive example of the extreme degree of structural modification that an ordinary segmented leg may take on. The newly hatched ostracod larva is in the nauplius stage of de- velopment (fig. 4 A), since it has only the three usual pairs of naupliar appendages. It is not a typical nauplius, however ; the antennae and mandibles are uniramous, and the body is already enclosed in a shell formed in the egg. Here is a good demonstration, then, that the crustacean nauplius, in addition to its primitive features, can take on a specialized structure characteristic of the order to which it belongs. During the postnaupliar stages, as shown in the series of drawings (fig. 4) here copied from Schreiber (1922) on the development of Cyprinotus incongruens, the postmandibular appendages are succes- sively added until the definitive number of seven in all is present in the eighth instar (F), in which the larva has attained essentially the adult structure. There is no question that the naupliar appendages are the anten- nules, the antennae, and the mandibles, but there has been some difference of opinion as to the identity of the postnaupliar appendages. In the Cypridae the first appendage after the mandible (fig. 5 B, 4) bears a large, flat, fringed lobe projecting upward in the shell cavity, and this appendage is commonly regarded as the maxilla. The next appendage (5) Schreiber termed the maxilliped. These two append- ages on each side in Cypris arise side by side on the arm of the hypo- stome (D, 4, 5), and Cannon (1926) regarded them as the maxillula and the maxilla, respectively. In Limnocythere inopinata (A), how- ever, as in other Cytheridae and in Nesideidae, appendage 5 is a typical leg well separated from 4. If, therefore, appendage 7 is interpreted as the maxilla, appendages 5, 6, and 7 are thoracic legs, and Kesling (1951) says this is now the accepted interpretation of 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 the ostracod limbs. Since we are here not particularly concerned with the identification of the appendages, they have simply been numbered on the drawings for purposes of comparison. Fic. 4.—Ostracoda. Larval stages of Cyprinotus incongruens Ramdohr (from Schreiber, 1922). A, nauplius, with three pairs of appendages. B, second instar, with one pair of added appendages (4) and caudal furca (f). C, fourth instar, with fifth ap- pendages (5). D, fifth instar, with sixth appendages (6). E, sixth instar, with seventh appendages (7). F, eighth instar, essentially adult structure. Inasmuch as in such forms as Limnocythere (fig. 5 A) the antennae and the last three pairs of appendages have the form of segmented legs, and in Cypris (B) the sixth and seventh appendages are typical legs (E), it may be inferred that the primitive ostracod appendages NOF LO CRUSTACEAN METAMORPHOSES—SNODGRASS 19 Fic. 5.—Ostracoda. A, Limnocythere inopinata (Baird), eighth instar (from Scheerer-Ostermyer, 1940). B, Cypris testudinaria Sharpe, adult, left shell removed. C, Philomedes globosa (Lilljeborg), adult, left shell removed. D, Cypris testudinaria, fourth and fifth appendages of left side and hypostome (Hst), posterior. E, same, sixth appendage. F, Philomedes globosa, fourth appendage. G, same, fifth appendage. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 were all ambulatory limbs. Moreover, since the legs of Limnocythere can be correlated with appendages of a very different structure in other genera, the ostracods give an actual demonstration of the changes that legs can undergo in adaptation to new uses. The very unleglike fifth appendage of Cypris (D, 5), for example, must have been evolved from a typical leg such as its representative in Limnocy- there (A, 5). This same appendage in Philomedes (C, 5) is again quite different from the corresponding appendage in Cypris. The maxilla of Cypris (D, 4) is represented in Philomedes (C, F, 4) by an appendage still suggestive of its leg origin. The fifth appendage of Philomedes (C, G, 5), however, has no resemblance to its counter- part in Limnocythere (A, 5), though the presence of three small marginal lobes and an apical tooth might be taken as evidence of a former segmentation. The sixth appendage (C, 6) has likewise three small marginal lobes and a broad, fringed apical lobe, but otherwise it has departed far from the structure of a leg (A,6). The seventh appendage of Philomedes (C, 7) has lost all semblance of a leg; it has become a long, flexible, vermiform cleaning organ armed with an apical brush of recurved bristles. The corresponding appendage in Cypris (B, 7)is likewise used for cleaning the shell chamber, but the only concession it has made to its function is an inversion of position. The sixth appendage of Cypris testudinaria (fig. 5 E) looks like a typical 7-segmented crustacean limb, counting the long terminal claw as the dactylopodite. The fourth and fifth podomeres of this appendage, however, are perhaps not true segments, since in Cypridop- sis vidua Kesling (1951, fig. 20) shows that the muscles from the sixth podomere have their origins in the base of the fourth podomere. The ostracods give no support to the theoretical phyllopod origin of crustacean limbs, and show clearly how simple segmented legs can be modified into very unleglike structures. COPEPODA The copepods include marine and fresh-water free-swimming spe- cies and a large number of parasitic species. They are nearly all very small crustaceans, mostly from 0.50 mm. to Io mm. in length in the adult stage. The simpler free-swimming copepods seem to approach more closely the typical shrimplike form of the higher crustaceans than do any of the other entomostracans. The body of a generalized form such as the marine Calanus (fig. 6) is divided into a cephalo- thoracic region bearing the appendages, and a slender limbless ab- domen. The cephalothorax includes an anterior unsegmented part NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 21 (7H) known as the head, or cephalosome, and a posterior thoracic region of five segments. The head carries the two pairs of antennae, the mandibles, two pairs of maxillae, and the first pair of legs, or maxillipeds. The five segments of the thoracic region bear each a pair of legs, but the legs of the last pair may be much reduced. The genital ducts open on the basal segment of the abdomen. The free-swimming copepods occur in such vast numbers in the ocean and in some inland lakes that they constitute a most important food source for many other aquatic animals from arrowworms to whales, but particularly for fishes. Being minute creatures themselves, the free copepods feed on the microscopic plant life of the water, 2Mx Mxpd' Fic. 6.—Copepoda. Calanus cristatus Kroyer, adult. TAnt, first antenna; 2Ant, second antenna; H, “head’; Md, mandible; 1M x, first maxilla; 2M-s, second maxilla; Mapd, maxilliped; VJ, VII, XI, body segments. which, elaborated in their own bodies, is thus passed on as food for the larger animals. It would seem, however, that the copepods have retaliated on the animals that eat them, since many species have be- come parasites of their potential enemies. Though fish are their favor- ite hosts, the parasitic copepods are not discriminative and attack almost every kind of creature that lives in the ocean. On the other hand, the copepods themselves are infested by numerous parasites, even by some of their own kind. Evidently life in the ocean is not a happy existence for either the predators or their victims. The nonparasitic copepods go through no changes of form in their life histories that can truly be called a metamorphosis. Their environ- ment is practically the same at all periods of their lives, and there is no call for adaptive modifications in either the larval or the adult stage. The successive developmental stages are merely steps in growth from youth to maturity. As an example, we may take the fresh-water 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I Cyclops and follow its life history as described by Dietrich (1915) and by Ziegelmayer (1925). The Cyclops larva hatches from the egg as a typical nauplius (figs. 7 A, 8A), which is followed by a second nauplius instar and four metanaupliar instars (fig. 8 B, C, D, E) in which the body lengthens, and finally (E) five pairs of appendages are added beyond the mandi- bles, including the maxillipeds (Mpd) and two pairs of legs (1L, Fic. 7—Copepoda. Nauplius and two metanaupliar instars of Cyclops (outlines from Ziegelmayer, 1925). A, nauplius, ventral, showing three pairs of appendages and labrum (Lm). B, first metanaupliar instar. C, fourth metanaupliar instar. 2L). At the next moult the larva (F) begins to take on the form and structure of the adult (G) and is now termed a copepodid, the ending id signifying that at this stage the larva has become copepodlike. The first copepodid acquires a third pair of legs; with further growth it passes through six copepodid instars until at last it becomes a sexually mature adult (G). Among Cyclops species there is thus no abrupt change between the various stages of growth, but new segments are added and the appendages develop from simple rudiments to their definitive forms. According to Ziegelmayer the segments formed after the nauplius stage are generated in a subterminal zone of growth. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 23 The genus Calanus, a typical free-swimming marine copepod of the suborder Gymnoplea, likewise develops from nauplius to adult by ordinary anamorphic growth without any metamorphic changes Fic. 8.—Copepoda: Podoplea. Developmental stages and adult of Cyclops. (A-F from Dietrich, 1915; G from Claus, 1863). A, Cyclops strenuus Fischer, nauplius, 0.119 mm. B, same, first metanauplius. C, same, second metanauplius. D, same, third metanauplius. E, same, fourth metanauplius. F, same, first copepodid, 0.303 mm. G, Cyclops coronatus Claus, adult female with eggs, 3.50 mm. 1Ant, first antenna; 2Ant, second antenna; 1L, 2L, 3L, legs; Md, mandible; IMz, 2Mx, first and second maxillae; Mxrpd, maxilliped. adaptive to different ways of living at different stages. The life history of the common Calanus finmarchicus has been described by Lebour (1916). The first six instars the author calls nauplii, but some of them would ordinarily be regarded as metanauplii, since two posterior segments and indications of a third segment are said to appear in the 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I fifth stage, with traces of the fifth and sixth pairs of appendages. In the sixth instar, judging from related species, there are present second maxillae, maxillipeds, and two pairs of swimming legs. The next instar is that of the first copepodid, which has three pairs of legs; the fourth legs appear in the second copepodid instar, and the definitive number of five is present in the third copepodid. The fifth and last copepodid is essentially like the adult. The free-swimming copepods, therefore, have a typical anamorphic development. Being crustaceans, they are primarily constructed for life in the water, and so long as they maintain a free existence there is no need of metamorphic adaptations to any other way of living. When now we turn to the parasitic copepods, the story is very different. An aquatic animal that hatches as a freely swimming larva and then becomes sedentary on another animal from which it extracts its food changes its environment and its mode of living in a very radical way. In some manner difficult to understand metamorphic changes of structure have been evolved that adapt the parasitic animal to its life of parasitism, and in many cases the transformation has been carried so far that the adult parasite could not be identified, or even recognized as a crustacean, if its early stages were not known. A few copepods appear to be transitional in their habits between a free life and one of parasitism. Such species are termed semiparasitic by Wilson (1921b), who says they are found on worms, mollusks, echinoderms, and in the gill chambers of crabs. These species are capable of swimming freely in the water, and their residence on any one host may be temporary. Their mouth parts, according to Wilson, are not suitable for either chewing or sucking and appear to be adapted for licking nourishment from the animals to which they attach them- selves. A species with biting mouth parts, however, could hardly resist sampling the blood of its host and then becoming an habitual parasite. The truly parasitic copepods include a large number of species, all of which undergo striking metamorphic adaptations to the nature of the host or the part of the host attacked, and some of them lead a double life on two different species of hosts. Some parasitic copepods undergo their metamorphoses during the larval development and become again free living in the adult stage; others remain on the host and attain their highest degree of metamorphosis as adults. Most of them, however, hatch from the eggs as typical nauplii, and in this stage or the following copepodid stage they must find their proper hosts. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 25 As an example of the life history of a parasitic copepod that returns to a free life in the adult stage, we may take the monstrillid Cymba- soma rigidum Thompson, described by Malaquin (1901) as Haemo- cera danae (Claparéde), which in its larval stages lives in the blood vessel of the serpulid worm Salmacina dysteri Huxley. The nauplius (fig. 9 A) has the usual three pairs of naupliar appendages, but the mandibles are recurved hooks, and the young larva has no mouth or alimentary canal. It is poorly fitted for swimming, and Malaquin suggests that the females probably sow their eggs over a colony of the serpulids. When in contact with a worm the nauplius attaches itself by its mandibular hooks to the worm’s integument, but it has no special organs for penetration. The skin of the worm, however, is delicate, and, a puncture once effected, the nauplius does a most surprising thing; it casts off its own cuticle and its appendages and forces its soft nude body into the host. Within the latter it becomes a shrunken, oval mass of undifferentiated cells (B), as if it had re- turned to an early embryonic condition to begin development all over again. In this form the parasite traverses the coelom of the host and makes its way into the ventral blood vessel. Here it secretes a new cuticle and then from its ventral side anteriorly there grow out two tapering, armlike processes (C) that extend posteriorly in the blood vessel of the worm and will serve the parasite as food-absorbing organs. Here, therefore, we see a metamorphic development adapting the parasite to its life in the host that certainly had no counterpart in the presumed free-living ancestors of its species. It is hard enough to believe the facts themselves, and we can speculate in vain as to how they all came about in evolution. The nauplius is prepared in advance for the life it is to lead by being provided with hooked mandibles, but what induces it to shed its cuticle and appendages and to squeeze itself into the worm? With the growth of the young larva in the worm (fig. 9H) the nutritive arms increase in length (D, E), the new cuticle is drawn out into a rostrum in front (E, Rk), and on the enlarged conical posterior part of the body it becomes armed with circles of spines directed forward. The organs of the future adult now gradually de- velop within the cuticle of the larva (F), and the abdomen forms as a ventral flexure (Ab) of the posterior part of the body. At an early stage the first antennae are regenerated (F, G, 1Ant) and eventually penetrate into the rostrum (1) when the head tissue has receded from the latter. From this point on the parasite develops normally into the adult form within the cuticular sheath of the larva. Finally, when its development is almost completed (1), the parasite becomes strongly fh + LU DWN bs LTE Fic. 9—Copepoda: Monstrillidae. Larval stages and adult of Haemocera danae (Claparéde) parasitic in the polychaete Salmacina dysteri Huxley (from Malaquin, 1901). A, free nauplius. B, after penetration into host. C, same, with nutritive ten- tacles (t). D, same, later stage. E, later stage, with cuticular envelope, rostrum (R), and spines. F, beginning transformation to adult inside cuticular sheath, rudiment of abdomen (Ab) bent forward. G, later stage of male, showing testis (Tes). H, specimen of Salmacina with two parasites in ventral blood vessel. I, male parasite almost adult. J, adult female, free after shedding the sheath and leaving the host; rAnt, first antenna, ef, egg filaments, gSeg, genital segment, Ov, ovary. 26 NO. ILO CRUSTACEAN METAMORPHOSES—SNODGRASS 27 active, doubling and straightening upon itself with the result that it ruptures both its enclosing sheath and the integument of the host. Then it escapes, leaving behind in its late host its spiny cuticle and its nutritive arms, which will no longer be needed. The monstrillid thus, according to Malaquin, makes during its life only two moults, one on entering the host, the other on leaving it. With its liberation the adult becomes at once an active free-swimming copepod (J). It now has only one pair of antennae and four pairs of swimming legs, and it lacks a complete alimentary canal. The body of the female, however, is mostly filled with a great mass of eggs (J, Ow) ; the busi- ness of the adult is the procreation of more parasites. Members of the family Caligidae, mostly parasitic on fish, are also free in the adult stage, but, though the adults are at liberty to leave the host and are equipped with swimming legs, they still depend for their food on the host that nourished them as larvae or on some other fish of the same kind. They, therefore, live largely as free external parasites. The structure and habits of many species of Caligidae have been described by Wilson (192Ia), and a detailed account of the larval stages of Caligus curtus (O. F. Miller) is given by Heegaard (1947). In Caligus curtus, according to Heegaard, there are two naupliar instars, the second of which goes over directly into a first copepodid without an intervening metanaupliar stage. The first copepodid is followed by a second copepodid, and then come five larval stages in a form known as a chalimus before the individual becomes adult. The actively swimming first copepodid has the responsibility of finding a host, which will be a codfish. It grasps a scale or a fin ray of the fish by means of its clawed second antennae, and holds on with the maxillipeds. After attachment the copepodid moults into the second copepodid (fig. 10 A). In this stage a gland in the head produces a secretion which will be discharged from the frontal region as a fila- ment (B), which becomes firmly fixed to a scale or a fin ray of the host. The parasite now becomes quiescent and takes no food as it hangs motionless on its attachment line, while within its cuticle a development takes place that will transform the copepodid into the first chalimus. This quiescent period of the copepod (B) is termed by Heegaard and some other writers a “pupa,” but, though motionless and nonfeeding, it is not comparable to the pupa of an insect. The insect pupa is a stage in itself during which the metamorphosed larva reverts to the parental form. Each larval instar of any arthropod begins its development within the loosened cuticle of the preceding instar. The copepod “pupa,” therefore, is merely the second copepodid 28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 after completion of its own growth when the first chalimus is de- veloping beneath its cuticle. An insect larva in a corresponding stage becomes quiescent and ceases to feed, but it is not a pupa. This con- cealed period in which any instar begins its development within the Fic. 10.—Copepoda. Larval and adult stages of fish parasites. (A, B from Heegaard, 1947; C, D from Wilson, 1905; E, F from Wilson, 1921a; G from Wilson, 1917.) A, Caligus curtus (O. F. Miiller), second copepodid. B, same, transforming stage of second copepodid. C, same, mature male chalimus. D, same, adult male. E, Trebius latifurcatus Wilson, adult female. F, Blakeanus corniger Wilson, adult female. G, Haemobaphes cyclopterina (Fab.), adult female. cuticle of the preceding instar has been termed by Hinton (1946) the pharate, or cloaked, phase of development. The young chalimus that emerges from the copepodid cuticle is not particularly different from the copepodid, though it is somewhat more advanced in development. Its first concern is to reattach itself to the NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 29 host, since the old filament remained with the discarded copepodid skin. Heegaard (1947) gives an interesting account of how the young chalimus with the pointed frontal lobe of its head bores a hole in a fin ray of the host. Into the wound thus formed is injected the secretion from the head gland, which hardens and holds fast, while the chalimus backs away and draws it out into a filament that secures the parasite to the host, but still allows it to move about on its tether. According to Heegaard each of the four succeeding chalimus stages reattaches itself in the same manner. The chalimus (C) was given its name because when first discovered it was thought to be the adult of an unknown species. Since the chalimus stages progressively de- velop from the second copepodid to the adult (D), they evidently represent the later copepodid stages of free-living copepods. The adults of Caligus curtus (fig. 10D) have pretty much the form and structure of an ordinary copepod, but, having no attachment to the host, both the males and the females are free to swim away. The egg-carrying female of another species with similar habits is shown at E of the figure. Since these copepods are dependent on a host for food in the adult stage, they retain their parasitic habits and are generally found crawling and feeding on the host, though they have not become specially modified in structure for a life of parasitism. This condition of dependence on a host, however, Wilson (1915) points out, constitutes the first step toward adult degeneration. If the adult parasite finds it advantageous to remain on the host, organs of locomotion become unnecessary, and in the end all that is needed are organs of nutrition and reproduction. The species shown at F, parasitic in an ascidian, still retains its appendages and a segmented abdomen, but the thorax has taken on a strange shape. The female at G, however, a permanent parasite on the gills of a fish, has de- generated from the copepod structure almost to the limit of simpli- fication. Yet, as already noted, “degeneration” is merely adaptation by the elimination of unnecessary organs. An example of an intermediate degree of degenerative simplification is seen in the lernaeopodid fish parasite Achtheres ambloplitis (fig. 11) described by Wilson (1911). In this copepod, Wilson says, the naupliar and metanaupliar stages are completed in the egg, and the larva hatches as a copepodid (A). During the egg stage the head gland produces a filament, which is still coiled in the head of the emerging copepodid (A, f). The young larva has two pairs of feath- ery swimming legs, and its maxillipeds (M-apd) are armed with strong hooks. It swims actively in search of a host, which must be a fish of the surface-swimming Centrarchidae. That the young copepod 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 deliberately presents itself to the head end of a fish has perhaps not been observed, but the fish unwittingly engulfs the copepod as food, which is of course just what the prospective parasite wanted it to do. To save itself from being swallowed the copepod grasps a gill arch of the fish with the hooks of its maxillipeds. Then it pushes its head into Fic. 11.—Copepoda: Lernaeopodidae. Developmental stages of fish parasites (from Wilson, 1910, 1915). A, Achtheres ambloplitis Kellicott, first copepodid, with filament (f) in head. B, same, second copepodid. C, same, with filament extruded and attached. D, same, adult male. E, same, adult female. F, Salminicola siscowat (Kellicott), egg-carrying female. the soft skin of a gill, which act breaks the cuticle over its head and releases the filament. The filament protrudes into the wound of the gill and the end spreads out into a disc that anchors the parasite inside the gill chamber of the fish. The first copepodid larva of Achtheres (fig. 11 A) undergoes a moult and enters a second copepodid instar (B), which is decreased in size and has taken on a different shape. The swimming legs, being now useless organs, are greatly reduced and later disappear (C). The mandibles have become toothed piercing organs for feeding. The NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS 31 large second maxillae (C, 2M) are much thickened and securely grasp the base of the attachment filament (f) by means of hooks imbedded in apical depressions. Then the larva backs away and draws the filament out to its full length, and thus maintains its hold on the gill with sufficient freedom of movement for feeding. At the next moult the sexes are mature. The female grows to a length of 4 or 5 mm., but the male remains a pygmy not over I mm. long. In the adult female (fig. 11 E) the maxillae are greatly lengthened, but the filament (f) is contracted so that only a short stalk projects beyond the maxillae. The maxillae of the male (D) are relatively not so long as those of the female, but the filament is unshortened. The filament, being a product of an internal head gland, is not shed and renewed at the moults; it retains its attachment and thus allows the parasite to complete its life in security within the gill chamber of the fish. The long filament of the small adult male permits the male to swing around on his tether until he comes in contact with a female, whom he grasps with his maxilliped claws and then lets go his hold on the filament, which remains attached to the gill. The female of another similar species of the genus Salminicola (F) is depicted by Wilson (1915) carrying her extruded eggs (es) in two long cylindrical sacs projecting from the gonopores while still attached to the gill of the fish. The newly hatched young presumably are carried out of the gill chamber in the expiratory currents of water. A good example of a parasitic copepod that inhabits two hosts dur- ing its life is the well-known fish parasite Lernaeocera branchialis (L.), a member of the Lernaeopodidae. This species during its larval life is an attached parasite on the gills of a flounder, but when adult both the male and the female become free and leave the flounder. The male undergoes no further transformation, and, after mating with a female still on the flounder, his purpose is accomplished. The female, on the other hand, is not yet sexually mature, and some instinct now urges her to leave the flounder and to seek a cod on which to com- plete the development of her ovaries. Once attached in the gill chamber of a cod she goes through an adult metamorphosis by which she is functionally reduced to the bare essentials necessary for feeding and egg production. For an account of the life history of Lernaeocera branchialis we may draw on the work of Pedaschenko (1898), Scott (1g0o1), Wilson (1917), Schuurmans-Stekhoven (1936), Sproston (1942), and Capart (1948). There is some difference of opinion concerning the nature of the early forms of this species. Pedaschenko says the first larva is a metanauplius (fig. 12 B) ; Scott and Sproston observed only one early 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I UAE GCPM ECU URAC AAC a AC RC CCC Fic. 12—Copepoda. Developmental stages of the fish parasite Lernaeocera branchialis (L.). (A, C, D, E, G from Sproston, 1942; B from Pedaschenko, 1898; F, I from Capart, 1948; H, J from Scott, 1901.) A, nauplius, 0.37 mm. B, metanauplius. C, free-swimming copepodid. D, third instar of chalimus stage on flounder. E, free-swimming adult male, 1.55 mm. F, young inseminated female on flounder. G, female on cod, beginning metamorphosis to penella stage. H, female in fully developed penella stage. 11.4 mm, I, adult egg-carrying female on cod, 40 mm. J, section of mature female. An, anus; AlCnl, alimentary canal; 1Ant, first antenna; 2Ant, second an- tenna; CmGld, cement gland; es, egg string; Mth, mouth; 2M, second maxilla; M-pd, maxilliped; Odct, oviduct; Ov, ovary; Sphr, spermatophore. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 33 stage, and called this stage a nauplius (A) ; while Wilson and Capart record both a nauplius and a metanauplius. The matter is of no particular importance for us in a study of the metamorphosis of the species. Whatever the larva that hatches from the egg may be, it moults into a free-swimming copepodid (C). Though the copepodid is only about half a millimeter in length, it has the responsibility of finding a flounder and of fixing itself to the gills of the fish, for which latter purpose it is provided with strongly chelate second antennae. Its hold on the gill, Sproston says, is never relinquished, and becomes the anchorage of the parasite until the free-swimming adult stage is reached. The gill filaments, however, are grasped also by the second maxillae in order to bring the mouth parts into close contact with the tissues on which the parasite feeds. When the copepodid moults the larva becomes a chalimus (fig. 12 D), but there is little change in form or structure. The chalimus, however, in its first instar acquires an additional attachment on the host in the form of a filament secreted by a gland in the head, which is anchored in the gill by two diverging branches that penetrate into punctures in the gill tissue. The rest of the secretion from the gland, Sproston says, falls back on the head of the larva where it hardens into a conical hood. The chalimus goes through four instars, and with each moult but the last a new hood is formed while the old ones remain, so that there are thus formed a set of overlapping caps corre- sponding in number with the moults. The third instar of the chalimus, to be identified as such by its three hoods, is illustrated at D of figure 12, redrawn from Sproston. The copepodid and the chalimus are metamorphic larval forms adapted to their respective functions of swimming and parasitic feeding. During its four instars the chalimus gradually approaches the adult structure, which is attained at the fourth moult after the copepodid stage. The adult male of Lernaeocera (fig. 12 E) leaves the old attachment filament with the castoff chalimus cuticle hanging on the gill of the flounder, and goes off in search of a female. The female (F), how- ever, awaits the coming of a male before she relinquishes her hold on the flounder. When the male finds a female still attached, mating takes place; two large spermatophores are inserted into the genital ducts of the female and are eventually lodged in her lengthened genital segment (F, Sphr). The female, still not sexually mature, then frees herself from the flounder and swims away to look for her second host, which should be a cod. On attaining a prospective victim, the female fixes herself to the bases of the gills by her second antennae, and now begins her metamorphosis into the final egg-producing stage. 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 After attachment on the cod the head of the female undergoes a curious transformation. Large branching, hornlike processes grow out from it and sink into the host tissue as anchoring devices (fig. 121). The proboscislike mouth region penetrates deeply in the flesh at the base of the gill until a large blood vessel is reached, from which the female will draw a rich nourishment for the maturing of her eggs. The first change of the body is a lengthening of the abdomen, princi- pally the genital segment, which grows out in a twisted wormlike form (G), and finally (H) becomes a long, straight, slender ap- pendage hanging from the thorax. At this stage the female is known as a penella from her resemblance to another adult copepod of that name. In the figure the penella stage shown at H is, of course, drawn on a much smaller scale than is the female at F or G. Next, the abdo- men swells into a great, elongate, twisted bag (1). The female in her final stage is said by Wilson (1917) to attain a length of 4o millimeters when fully extended. From now on she is merely an egg- producing organism. Her internal organs (J) consist principally of the enlarged alimentary canal (AlCnl), the ovaries (Ov) and oviducts (Odct), and a pair of cement glands (CmGld) that form the casings for the eggs. The eggs are discharged in two long coiled strings (es), which, Wilson says, reach a lenth of 150 to 200 millimeters. Consider- ing the number of eggs that the species produces, any flounder or cod may consider itself lucky if it escapes infestation. According to Schuurmans-Stekhoven there is only one generation of the parasite each year. The metamorphosis of Lernaeocera branchialis affects principally the female on the secondary host. The less modified chalimus instars carry on the developmental processes while attached on the flounder as do the copepodid stages of free-living species. The adult male and the adult female on leaving the flounder are normal, swimming cope- pods. The transformation of the female on the cod involves, on the one hand, a simplification of the thorax until it becomes indistinguish- able from the abdomen, except for the retention of the appendages ; but, on the other hand, there is a new development of anchoring process on the head, and a great overgrowth of the reproductive part of the body. The metamorphosis of the female, therefore, is both recessive and progressive in an anatomcial sense. A study of the de- velopment and metamorphosis should take into consideration not only the anatomical changes that the individual goes through, but also the changes in its instincts. The copepodid of Lernaeocera, for example, must have an instinctive urge to attach itself to a flounder; the adult female instinctively leaves the flounder and looks for a cod. NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS~ . 35 Copepod fish parasites are not all content with attacking the scales, fins, or gills of the host. Some make their abode in the nostrils of the fish; others penetrate through the skin into the body cavity where they attack the vital inner organs. The worst of them are members of the genus Phrixocephalus, several species of which are described by Wilson (1917). These parasites bore into the eyes of their victims in order to feed from blood vessels at the back of the organs. Para- sites seem to have been endowed by nature with great versatility, but the life of a fish is nothing to be envied. CIRRIPEDIA The cirripeds include the familiar barnacles and several groups of parasitic species. The first-stage larvae in most cases are nauplii usually characterized by the presence of a pair of lateral frontal horns on the anterior part of the body. In some species the horns are merely short spines (figs. 14 B, 16 A, fh), in others they are long and either straight or curved, but when present the horns identify the nauplius as a young cirriped. The nauplius becomes a metanauplius; the meta- nauplius transforms into a free-swimming larval stage known as a cypris because its body is enclosed in a bivalve shell with a closing muscle, and thus resembles the ostracod of the same name. The cirriped cypris (fig. 14 C) has six pairs of swimming legs, a simple median eye, compound lateral eyes, and a pair of antennules project- ing from the anterior end of the shell. After swimming freely for some time the cypris of most species attaches itself by the antennules to some solid object on which it remains permanently fixed and here develops into the adult form. The barnacles in the adult stage (fig. 14 F, H) are sedentary on rocks, clam shells, wooden piles, ship bottoms, whales, or almost any- thing else in the ocean, and they get their food from the water. The parasitic cirripeds attach themselves to other animals and derive their sustenance from the host. The adult barnacles retain enough of their ancestral structure to be recognized as crustaceans ; some of the para- sitic cirripeds, on the other hand, undergo such extreme degrees of adult metamorphosis that their crustacean derivation is known only from their early larval stages. The Ascothoracica.—The members of this suborder are of particu- lar interest because as adults they appear to be equivalent to the cypris stage of other cirripeds. If they truly are cirripeds, therefore, they evidently are a primitive group of the order, and suggest that the cirripeds have been derived from cyprislike ancestors, perhaps re- 36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 lated to the bivalved Ostracoda. From the standpoint of metamorpho- sis the Ascothoracica are of small interest, since whatever modifica- tions some of them do undergo effect principally a simplification of the cypris structure. They are all minute creatures parasitic on Actinozoa and Echinodermata. The least modified member of the Ascothoracica is Synagoga mira Norman (fig. 13 A), which lives externally on the black corals Antipathes, clinging to the host by the large antennules. Since Syna- goga has well-developed setigerous legs, however, it appears probable that it can relax its hold and swim from one host to another. The species is known only from a few specimens described by Norman (1913). The head and thorax are enclosed in a large, oval bivalve shell, 4 millimeters in length, provided with strong adductor muscles, but the slender, five-segmented abdomen projects freely from the shell and bears a pair of long uropods. The large antennules (7Ant) are armed with apical hooks; the six pairs of thoracic legs bear long setae and are evidently adapted for swimming. The mouth parts as described by Norman are slender piercing organs enclosed in a large conical proboscis (Prb). Of all the Ascothoracica, Synagoga mira alone appears to have no metamorphosis and to have retained the ability to swim; no other species, therefore, has so good a claim to being a primitive cirriped. A related member of the Ascothoracica is described by Okada (1938) as Synagoga metacrinicola (fig. 13 B). This species has the entire body enclosed in the shell, the abdomen being relatively short, but otherwise it is similar to S. mira. Okada finds well-differentiated males and females in S. metacrinicola, the sexes being separate in most of the Ascothoracica, in which the males are much smaller than the females. He reports that Norman’s specimens, supposed to be females, are found on reexamination by sections to be males with mature spermatozoa. Okada thus demonstrates that the known examples of Synagoga are adult forms and not larvae of an otherwise unknown species, as some writers had suggested they might be. The other Ascothoracica that are parasitic on horny corals appear as small budlike bodies on the coral stems. The shells are of various shapes and in some species are enclosed in a tunic derived from the host. In most of these forms the legs are more or less reduced and lack swimming setae. An ascothoracid described by Heegaard (1951) as Ascothorax bulbosa, found in specimens of an ophiuroid, or brittle starfish, has an oval shell (fig. 13 C), the small males being attached dorsally on the females beneath the cuticle of the latter. The body NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 37 of the animal (D, E) is somewhat deformed and the thoracic legs are reduced. The greatest modification among the Ascothoracica occurs in the Fic. 13.—Cirripedia: Ascothoracica. (A from Norman, 1913; B from Okada, 1938; C, D, E from Heegaard, 1951; F, G from Knipowitsch, 1890.) A, Synagoga mira Norman, adult. B, Synagoga metacrinicola Okada. C, Ascothorax bulbosa Heegaard, shell of female with small male on top, internal parasite of ophiuroid. D, same, female. E, same, male. F, Dendrogaster asteri- cola Knipowitsch, cypris larva. G, same, adult enclosed in branched mantle, internal parasite of starfish. Dendrogasteridae, which are internal parasites of echinoderms. Den- drogaster astericola, described by Knipowitsch (1890), is enclosed in a voluminous mantle (fig. 13 G) with large lateral lobes, which are penetrated by diverticula of the stomach. The cypris larva (F), 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 however, is a typical cirriped cypris, much resembling the adult of Synagoga (A). A species figured by Fisher (1911) as Dendrogaster arbusculus, found in a Californian starfish, has an elaborately branched structure. The known nauplii of the Ascothoracica, according to Okada, differ from the nauplii of other cirripeds in the absence of the usual frontal horns, another feature that sets the ascothoracicans off as a primitive branch of the cirripeds. Some species hatch as nauplii, others as meta- nauplii, and still others in the cypris stage. The Thoracica.—To this suborder belong the barnacles, which in the adult stage are enclosed in calcareous shells. Some are conical and sit flat on the substrate (fig. 14 F), others are flattened and sup- ported on stalks (H). When either kind is broken open, however, there is exposed within the shell a shrimplike creature (G) lying on its back or standing on its head with its cirruslike feet, when active, sticking out of the top or side (H) with a waving movement. The nauplius of a common barnacle such as Balanus, described by Runnstrém (1924-1925), has the typical naupliar structure (fig. 14 A) except for the pair of small horns (B, fh) on the anterior part of its body. Runnstrom describes two naupliar stages, but since the sec- ond becomes elongate and acquires rudiments of three postmandibular appendages it would ordinarily be called a metanauplius. After a few hours of swimming, the metanauplius abruptly transforms into a cypris (C) with a bivalve shell and long seta-bearing legs, wherewith it is better equipped for a pelagic life. Eventually the cypris fixes itself to a support by its first antennae (1Ant), each of which (E) is provided with an adhesive cup on the third segment. A cementing substance discharged through the antennae from glands in the head gives the cypris a permanent attachment. Then the cypris withdraws the hind part of its body and its legs into the shell, and now begins the formation of the plates of the adult barnacle. According to Runnstrom, the plates are first formed as chitinizations of the mantle and only later become calcified. When the plates have the essential adult pattern (D) the cypris shell is cast off, and with the moult the legs of the cypris are replaced by the cirri of the barnacle. The metamorphosis of the cypris into the barnacle is not excessive. It is a structural adaptation to the permanently sessile condition within the shell, and the eyes are absorbed as now useless organs. The changes that take place in the body have been described by Doochin (1951). The shell-closing muscle of the cypris is retained (fig. 14 G, mcl), and the mantle supporting the plates of the shell is attached to NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS 39 the body only around the ends of the muscle. The peduncle of the stalked barnacles is a product of the head and becomes occupied by connective tissue and muscles. The barnacles are hermaphroditic, but they generally live in crowded colonies and cross fertilization is made possible by a long, tubular penis arising at the base of the vestigial abdomen. The Rhizocephala.—In this suborder of parasitic cirripeds we en- counter the strangest metamorphic phenomena known in the whole Fic. 14.—Cirripedia: Thoracica. (A-E from Runnstrom, 1924-1925.) A, Balanus balanoides (L.), nauplius. B, same, anterior end of body with median eye and frontal horns (fh). C, same, cypris larva. D, same, later stage, barnacle plates formed inside cypris shell. E, same, first antenna of cypris with attachment cup on third segment. F, Balanus eburneus Gould, group of adults. G, Lepas anserifera L., adult animal in natural position removed from shell. H, same, stalked shell. animal kingdom. The rhizocephalans include a number of genera, of which the best known are crab parasites of the genus Sacculina. The visible external evidence that a crab is parasitized by a sacculinid is the presence of a large saclike body attached ventrally on the crab at the base of the abdomen (fig. 15 A). This external sac is the re- productive part of the parasite containing the ovaries and the testes, but from it long, rootlike processes extend into the body of the crab and serve for the nutrition of the parasite. The eggs are fertilized and hatch within the external sac, giving rise to nauplii, which trans- form into typical cirriped cypris larvae. The free-swimming cypris 40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I larvae escape through a hole in the sac, find another crab, and enter the latter after undergoing extraordinary transformation processes. The life history of Sacculina carcini Thompson was fully described and illustrated by Delage in 1884, and Delage’s account has been veri- fied, at least in part, by G. Smith (1906) and Veillet (1945). It still remains as the authentic history of a Sacculina, and the following story of the life and metamorphosis of this parasite is based on the papers by Delage and Smith, with illustrations taken from both. The nauplius of Sacculina (fig. 16 A) has the characteristic frontal horns (fh) of cirriped nauplii, but it lacks an alimentary canal and has neither a mouth nor an anus. After several moults the nauplius becomes a cypris larva (B) with a length of 0.20 mm. On leaving Fic. 15.—Cirripedia: Rhizocephala. External parasitic stages on crabs. A, Loxothylacus texanus Boschma, a sacculinid on Callinectes sapidus Rath- bun. B, Thompsonia on Thalamita prymna (Herbst) (from Potts, 1915). the brood sac on the crab, the cypris leads a free life in the ocean for several days. Finally, on finding a young crab that has just moulted, it attaches itself to the latter by one of its antennules (C, rAnt), which are provided with small suction cups. The point of attachment is usually in the membrane at the base of a hair (Hr). When firmly secured the cypris begins violent swinging movements of the body, which detach the thorax (7h) along with the legs and the abdomen and throw the whole rear part of the body out of the shell (Sh). From the large hole thus left in the head end of the cypris are now expelled most of the internal tissues, leaving only a mass of cells con- taining the reproductive elements. Later the hole closes. While this process of elimination has been going on, other changes take place. The body of the larva separates from the shell (D) and contracts to a sac walled by the ectoderm, which is much smaller than NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 4I the original cypris (B). Soon a new cuticle is secreted on the surface of the sac (D, rCt) in continuity with the cuticle of the attached antenna, and the larva becomes a compact oval body still within the shell but now entirely free from it. Again, as if preparing for a moult, a second cuticle (2Ct) is formed beneath the outer one, and a small point (d) grows out from its anterior end into the hollow of the antenna. The body of the larva then retracts within the outer cuticle (E), and as it does so the cuticular point elongates into a long, hollow dartlike tube (d) with the narrow end cut off obliquely like the point of a hypodermic needle, and its widened base embedded in the body of the retracted tissue of the larva. This newly formed organ Delage called the dart, and the larva armed with the dart he termed a kentro- gon (from Greek kentron, a dart, and gonos, a larva). The shell together with its loose inclusions is now thrown off, leaving the kentro- gon, still enclosed in the outer cuticle, attached to the erab by the antenna (F). The body of the larva again expands and pushes the dart into the antenna (fig. 16 F) until its tip comes into contact with the integument of the crab. Since the parasite is held fast by the antenna, the dart pierces the integument instead, pushing the larva away from it, and finally (G) projects into the body of the crab. Now the soft tissues of the larva contract away from the cuticle but remain still connected with the base of the dart. The remains of the larva thus have a free passageway into the body of the crab through the narrow channel of the dart, the orifice of which is said by Delage to be 3 to 6 microns in diameter. Though Delage says he did not observe the actual passage of the larval substance through the dart, globules are seen inside the dart and the parasite is next found inside the crab. By the method of the Sacculina a mouse might get into the pantry through the keyhole of the door, but once inside it would have to devise a new way of eating. This problem the Sacculina solves very easily—it simply adopts the feeding method of a plant by sending out absorbent roots among the organs of the crab. Inside the crab the parasite becomes a small oval body consisting of a mass of cells enclosed in an ectodermal epithelium. It finds its way to the ventral side of the crab’s intestine and here becomes at- tached. Now the principal concern of the parasite is to obtain nourish- ment from the host for maturing the germ cells which it has brought with it from the cypris stage. Incidentally, this will be the first food from an external source that the larva itself has had, since it was hatched without an alimentary canal. The larval body expands against the intestine of the crab (fig. 17 A) and sends out branching 42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 Fic. 16.—Cirripedia: Rhizocephala. Metamorphosis of Sacculina carcint Thomp- son (from Delage, 1884). A, nauplius. B, free-swimming cypris stage ready to moult. C, cypris fixed by antenna at base of hair of crab, shell separated, thorax detached and thrown off with internal tissues. D, larva still in shell has formed a new cuticle (1Ct). E, shell being shed, larval body retracted within cuticle, with long, hollow “dart” (d) extended toward base of antenna. F, larva with a second inner cuticle (2Ct), the dart extended into antenna. G, larval body expanded, the dart has pierced the hair membrane of the crab. Ab, abdomen; rAnt, first antenna; rCt, outer cuticle; 2Ct, inner cuticle; d, dart; fh, frontal horn; Hr, hair of crab; Sh, cypris shell; Th, thorax. NO; LO CRUSTACEAN METAMORPHOSES—SNODGRASS 43 rootlike processes (rhizai), which continue to grow, branch, and unite until a network surrounds the intestine (B), from which branches penetrate between the other organs and extend out into the append- ages. The roots do not enter the tissue of the crab, but Delage says only the heart and the gills are not attacked. These are the organs necessary for maintaining the life of the host and therefore that of the parasite, but how did the parasite ever learn to discriminate? The Sacculina at this stage has been aptly likened to a fungus. That a crustacean can be so transformed shows the unlimited potentialities of metamorphosis. When nutrition has been fully provided for, attention must be given to the reproductive function. If the eggs were allowed to hatch inside the crab, the young larvae would find themselves in a prison from which there would be no escape. The body of the parasite, therefore, emerges through the ventral integument of the crab and becomes a brood chamber in which the eggs mature and from which the larvae are liberated into the ocean. The pressure of the parasite’s body causes a dissolution of the crab’s epidermis beneath it, and prevents the formation of cuticle at this point. Consequently at the next moult of the crab the Sacculina body containing the reproductive cells emerges and becomes external, but is still connected with the crab by a short peduncle giving passage to the feeding roots. The place of emergence is at the middle of an abdominal segment ; if it were inter- segmental, movements of the abdomen might constrict the peduncle and shut off the food supply of the parasite. It seems that the simpler a creature may be in its organization, the more does nature guard it against emergencies. It is interesting to note that the species shown at A of figure 15 is exactly modeled to fit into the pocket between the under surface of the thorax of the crab and the reflexed abdomen beneath it. The external parasite, as seen in section (fig. 17 C, D) consists of a central mass of cells contained in a tunic suspended from the peduncle, and of an outer mantle (mn) that encloses a peripheral brood chamber (bc). The figures at C and D, taken from G. Smith (1906), depict a species of Peltogaster, but the structure is essentially the same in Sacculina. The cells of the central mass are the eggs in the ovary (Ov) ; above them is a pair of tubular testes (Tes) anda single nerve ganglion (Gung). The ripe eggs are discharged into the mantle cavity and here fertilized by spermatozoa from the testes, the parasites being necessarily hermaphroditic. The larvae escape in the cypris stage from an opening (D, op) in one end of the brood chamber. Successive lots of eggs are discharged and fertilized, and after each brood of larvae the cuticular lining of the brood chamber 44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 is shed. The maturation of the sperm and the eggs and the fertiliza- tion of the latter are fully described by Smith, but here ends our discussion of the metamorphosis of Sacculina. Fic. 17.—Cirripedia; Rhizocephala. Internal and external parasitic stages. (A-D from G. Smith, 1906; E, F from Potts, rors.) A, Sacculina neglecta attached on intestine of crab Inachus scorpio. B, same, later stage with root system developed. C, Peltogaster sp., diagrammatic cross sction of parasite after emergence on ventral side of crab. D, same, longitudinal section. E, Thompsonia sp., part of root system in tail fan of crab Synalpheus brucei, with external brood sacs. F, same, external sacs on chela of Thalamita prymna. a, internal reproductive buds; b, external brood sac containing cypris larvae; be, brood cavity; c, external sac with all but a few larvae escaped through terminal aperture (op) ; Gng, ganglion; mn, mantle; Od, oviduct; of, external opening of brood cavity; Ov, ovary; r, nutritive roots; Jes, testis. The parasitization of the crab by Sacculina adversely affects the gonads and results in structural changes of the host called parasitic castration. At the moult accompanying the emergence of the parasite, the male crab takes on certain female characters and the female suffers NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 45 a change from normal. Inasmuch as Sacculina produces only one re- productive body, the parasite has no concern with what happens to the host. Peltogaster socialis, another rhizocephalan, differs from species of Sacculina in that a number of parasites, 2 to 30 of them, all in about the same stage of development, are found on the outside of one host. In his investigation of this species, G. Smith (1906) reported that each external parasite appeared to have its individual root system in the crab. Potts (1915') questioned the accuracy of Smith’s observa- tion, and suggested that more probably the several external parasites arise from a common root system, pointing out that Peltogaster socialis is a comparatively rare species and that it would seem unlikely that so many cypris larvae should attack the same crab at the same time. That many external reproductive sacs may arise from one internal system of roots has been amply demonstrated by Potts (1915) in his study of the genus Thompsonia. Species of this genus, parasitic on various crabs, reach the ultimate in the conversion of an adult crustacean to the status of a fungus. The parasite within the host has the form of an extensive and intricate network of fine branching and anastomosing threads distributed principally on the ventral wall of the abdomen at both sides of the nerve cord, but also entering the thorax where the branches may extend up on the lateral and dorsal walls. The root threads, according to Potts, are from 10 to 20 microns in thickness. From the central network branches penetrate into the thoracic and abdominal appendages and into the lobes of the tail fan. On the branches in the appendages are developed small budlike processes (fig. 17 E,a) that project outward against the integument. These buds contain the germ cells that will become ova. At the next moult of the crab they break through the soft new cuticle and become small external sacs (E, b, and F) standing on the surface. The sacs may be so numerous that the appendages, especially the legs, are loaded with them (fig. 15 B). These external sacs are the reproductive organs of the parasite, and might be likened to the spore-bearing bodies of a fungus nycellium. Since Thompsonia produces no male elements, the eggs are apparently parthenogenetic. They hatch di- rectly into young cypris larvae (fig. 17 E), which, before the next moult of the crab, escape from the sac through an apical perforation (op). The empty sacs are carried off on the exuviae at the following moult of the crab. The development of the eggs, therefore, is so regu- lated that the larvae reach maturity during the time between moults of the host. At each moult a new crop of egg sacs breaks out on the 46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 surface. Thompsonia, unlike Sacculina, appears to do no specific damage to its host, so that it can continue its parasitic life and indefi- nitely repeat its reproductive processes. The inoculation of the host by the free-swimming cypris has not been observed. The Thompsonia-infested crab presents one of the most curious anomalies in the whole realm of nature. Here are two crustaceans, one inside the other, the crab a highly developed arthropod, the para- site, a crustacean relative of the crab, spread out inside the latter in the form of a network of filaments. Both host and parasite are adult animals, each being the reproductive stage of its species. Progressive and regressive evolution could hardly reach a greater degree of divergence. Thompsonia is known to be a crustacean because it produces free- swimming cypris larvae, it is known to be a rhizocephalan because of its likeness to Sacculina, and Sacculina is known to be a cirriped because of the character of its nauplius. The barnacles and the rhizocephalans have in common the habit of attaching themselves to a support by the antennules in the cypris stage. From this point on they widely diverge. It would be highly interesting to know how the Sacculina larva learned to attach itself at the base of a hair on a crab, how it acquired the urge to get into the crab, and how it ever de- veloped a self-reducing method for doing it. Halfway measures would be useless. Clearly there are problems in evolution for which natural selection does not offer a ready solution. ISOPODA Most of the Malacostraca are too large to be parasites. The ma- jority are predatory, and few of them exhibit any considerable degree of metamorphosis. Most of them, moreover, hatch at a later period of development than do the Entomostraca, and some of them are al- most completely epimorphic. A prominent exception to the general free mode of life, however, occurs among the isopods, a few species of which have adopted parasitism, and have become structurally adapted to a parasitic life in a degree equal to that of some of the entomostracans. This fact shows how readily metamorphosis can crop out independently in species that have adopted a new way of living. The isopods in general are a conservative group in which the young hatch at a late stage of development with complete body segmentation and most of the appendages present. Among those that have become parasitic, however, varying degrees of adaptive metamorphosis occur NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 47 in the life history. Species that feed temporarily on the host only during the larval stages may undergo but little structural adaptation. On the other hand, species that are permanently parasitic are likely to go through a high degree of metamorphosis both in the larval and the adult stages. The two species described in the following pages, one belonging to the Gnathiidea, the other to the Epicaridea, may be taken to illustrate the two extremes of parasitic metamorphosis found among the isopods. Paragnathia formica (Hesse).—This isopod, parasitic in its larval stage on fishes, gives us a good example of a parasite that undergoes but a minimum of metamorphic adaptation to life on its host. The developmental life history of Paragnathia formica has been amply described by Monod (1926) and the following account with accom- panying illustrations (fig. 18) is taken from Monod’s work. The adult males and females live together in small burrows exca- vated in semihard mud banks of stillwater estuaries below the mean level of the ocean. Here the pregnant females in late summer or early fall give birth to active larvae. The newborn larvae leave the burrow, swimming with great speed by movements of the abdomen. Once in the open water they lose no time in attaching themselves to a fish; most any fish will do. The time between birth and attachment is a period of dispersal, during which the larva takes no food, subsisting on the remains of yolk in its alimentary canal. The larva attaches itself on the fish with its second maxillipeds, and the attack is made at any place that will readily yield blood, such as the membrane between the rays of a fin, the gills, or the mouth. The swimming larva (fig. 18 A) is a fully segmented young isopod with large compound eyes and a complete equipment of appendages. In its embryonic development it has been provided in advance with efficient piercing mouth parts and a sucking apparatus. The mouth parts (G) are enclosed in a large conical proboscis projecting forward from the head, formed of the epistome (Epst) above and the first maxillipeds (rMxpd) below. The long, strongly toothed mandibles (Md) are but little movable ; they serve as harpoons to hold the para- site close to the fish while the sharp-pointed, freely movable first maxillae (1M), supported by the paragnaths (Pgn) beneath them, puncture the integument. The much reduced second maxillae (2/7) have no recognized function in feeding. When the young larva (fig. 18 A) has once established itself on a fish and has begun to feed on the blood of the host, its form changes ; the change is said by Monod to be effected without the intervention of a moult. The thorax lengthens, accompanied by a swelling of the 48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 x ‘ ‘ Pgn. 1Mxpd 1Mx SMx ( Fic. 18.—Isopoda: Gnathiidea. Paragnatha formica (Hesse) (from Monod, 1926). A, first free-swimming larva. B, second form of larva parasitic on a fish. C, attitude of feeding larva. D, adult male. E, adult female. F, head of adult male, dorsal. G, section of larval head showing piercing mouth parts. Epst, epistome; Md, mandible; 1M+x, first maxilla; 2Mx, second maxilla; M-xpd, maxilliped; Pgn, paragnath. NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS 49 last three segments (B) ; the segmental limits disappear owing to the unfolding of the previously deeply infolded intersegmental mem- branes. This is the feeding stage of the parasite (B, C), called the pranize by Monod (L., prandium, lunch). Its meal lasts about six months. At the end of winter or the beginning of spring the fully fed para- sites leave the host and return to the bank of the estuary. The males individually dig burrows or take possession of empty ones in advance of the coming of the females. The completed burrows are 1.5 to 2.5 cm. in depth, sloping downward from the mouth to an inner chamber 4 or 5 mm. in diameter. When the females arrive they enter burrows already inhabited by a male; as many as IO or more may consort with a single male. Within the burrows both the male and the females undergo their first and only moult, accompanied by a small degree of metamorphosis. The cuticle splits crosswise over the thorax, and the two ends are cast off separately. The sexes are now differentiated and the isopods enter their third functional stage, which is that of reproduction. The male (fig. 18D) retains a relatively slender figure, but the female (E) becomes greatly distended with the development of the ovaries. The mouth parts of both sexes are reduced, except the mandibles of the male (F), which are long prongs perhaps used for digging or for holding the female in mating. Sub- sistence is now at the expense of the food consumed during the para- sitic stage. The eggs develop into mature larvae within the ovaries of the fe- male, which become distended into a pair of large, saclike uteri, com- pressing the empty alimentary canal between them. On the ventral surface of the female’s thorax are several pairs of small overlapping oostegite plates, and above them is a large atrial cavity, into which the oviducts open, but this cavity does not serve as a brood chamber. When the young issue from the uteri through the oviducts into the atrium, the oostegites open and the larvae precipitate themselves head first through the aperture directly into the water, where they at once begin active swimming. After giving birth to the young, the females quickly die, but the males are longer lived and their metamorphosis is not so closely correlated with the season. There is clearly in the life history of Paragnathia formica little that can be called a true metamorphosis. The change of form between the two larval phases is merely a distention and elongation of the thorax resulting from the unfolding of the intersegmental membranes. The metamorphosis at the moult to the adult stage involves principally a reduction of the mouth parts which are no longer used for feeding. 50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 Since most isopods have biting and chewing mouth parts, the con- version of the mouth parts in the embryo of Paragnathia into piercing organs may be regarded as an embryonic metamorphosis preparing the future larva for its prospective life as a parasite. Danalia curvata Fraisse-——This isopod belongs to the suborder Epicaridea, the members of which are parasitic on other crustaceans. It gives us an example of the sex versatility of some of the epicaride- ans in which the animal is first a functional male and then a functional female. In its female role Danalia curvata attaches itself to a crab infested with a rhizocephalan and feeds either on this parasite in its external state or on its roots in the host. Here the female is in- seminated by a young free-swimming male, after which the male attaches to the crab and becomes a female. In this manner, though the species is hermaphroditic, it avoids self-fertilization. The follow- ing outline of the life history of Danalia curvata is taken from the work of G. Smith (1906) and of Caullery (1908). The mature female (fig. 19 H) has no likeness whatever to a crustacean; she is little more than a sac full of eggs attached to the crab by a narrow stalk inserted into the crab’s body. The young on hatching leave the brood pouch of the mother and become free-swim- ming larvae. At this stage the larva (A) is recognizable as an im- mature isopod, and is called a microniscus. The larva is distinctly segmented, has two pairs of antennae, five pairs of thoracic append- ages, and five pairs of pleopods, but eyes are absent and the mouth parts are reduced to a pair of styliform mandibles enclosed in a small buccal cone. The microniscus larva may adopt a copepod as a tempo- rary host, as do most of its relatives. After several moults it takes on a different form (B, C) and is now termed a cryptoniscus, pre- sumably because its isopod characters are less evident. The body is more elongate and eyes have been developed, the appendages are retained ; the cryptoniscus is a free-swimming stage. Within its body is a pair of large hermaphroditic sex organs (B), each of which con- tains in its anterior end a small ovary (Ov) and in its posterior part a large testis (Tes). The testes rapidly develop and become filled with an abundance of spermatozoa. The larva is now a functional male. The male cryptoniscus seeks out a crab parasitized by a sexually mature female of his own species (G). After accomplishing the insemination of the female the larval male attaches himself to the crab or to the Sacculina on the crab by the first two pairs of his chelate pereiopods. Then a moult takes place, the cuticle being shed in two pieces from the opposite ends of the body, and it is then seen that the larva has undergone a radical change of structure within the NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 51 cryptoniscus cuticle. The body has become a small cylindrical sac (D) about one and a quarter millimeters in length in which all trace of segmentation has disappeared. The eyes are gone, and all the Fic. 19.—Isopoda: Epicaridea. Life history of Danalia curvata Fraisse. (A, C-H from Caullery, 1908; B from G. Smith, 1906.) A, first instar larva. B, second free larval stage, with hermaphroditic sex organs containing small ovaries and large testes. C, larva with testes fully developed. D, parasitic larva on crab. E, same, with proboscis elongated. F, functional female stage, with testes degenerated, ovaries fully developed. G, female containing brood sac. H, female in final stage. Mth, mouth; Ov, ovary; Prb, proboscis; 2Prpd, second pereiopod; Tes, testis. appendages have been cast off with the exuviae except a pair of small hooklike second pereiopods (2Prpd) with which the parasite main- tains its hold on the host. A small conical proboscis (Prb) bears the mouth on its end. The testes, now that they have performed their 52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I function, degenerate and the ovaries begin to develop, so that the former male larva thus changes functionally to a female. As a female, the parasite begins to increase in size. First the proboscis elongates (fig. 19 E) and, though it is armed with only a pair of minute teeth, it penetrates the integument of the crab. Inside the host the proboscis stretches out to a long neck (F, Prb) until the mouth (Mth) at the end comes in contact with the roots of the Sacculina, and four diverging processes grow out around the mouth to anchor the proboscis in the tissues of the crab. The body of the newly feminized individual then takes on a saclike form (G). The ovaries (Ov) are now fully developed; the oviducts open on two pairs of ventral papillae. At this stage the female is inseminated by a cryptoniscus larva still in the male phase of development (C). The fertilized eggs are discharged into a large incubation chamber be- neath the cuticle of the female. The process of forming the chamber is somewhat complex as described by Caullery, but essentially it ap- pears that two lateral ingrowths of the ventral ectoderm extend inward around the sides of the body, and eventually close over the orifices of the oviducts. When the eggs are discharged into the incu- bation chamber, the female ceases to feed, doubles on herself in the form of a U (H) and becomes a mere inert sac in which the eggs complete their development. EUPHAUSIACEA The Euphausiacea and some of the Penaeidae are exceptional among the Malacostraca in that they are hatched as nauplii. They are both marine and entirely pelagic. The euphausiids go through many moults before reaching the adult stage. Students of the group com- monly distinguish five immature stages in the life history of an indi- vidual. The first two are the nauplius and the metanauplius, the fol- lowing three stages are termed the calyptopis, the furcillia, and the cyrtopia. These forms, however, are merely stages of growth charac- terized by different degrees of differentiation toward the adult struc- ture (fig. 20 A-G). Except for the successive specialization of differ- ent groups of appendages for swimming there are few metamorphic changes involved in the development. The following condensed ac- count of the typical life history of a euphausiid species is based on the papers by Heegaard (1948) and Lebour (1925), with illustrations taken from both. The order includes only a single family, the Euphausiidae. The newly hatched euphausiid larva (fig. 20 A) is a typical nauplius of simple form with the usual three pairs of appendages, a simple NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 53 median eye, and a large labrum. The metanauplius (B) acquires three additional pairs of appendages, which are the first and second maxillae and the first pair of legs (rL), or maxillipeds. The mandibles (Md) we ite Fic. 20.—Euphausiacea. Life-history stages. (A-E from Heegaard, 1948; F, G from Lebour, 1925.) A, Meganyctiphanes norvegica Sars, nauplius. B, same, metanauplius. C, same, first calyptopis instar. D, same, third calyptopis instar. E, same, first furcillia instar. F, Nyctiphanes couchii Bell, last (12th) furcillia instar. G, Meganyctiphanes norvegica, first cyrtopia instar. have become jawlike. The metanauplius is followed by the calyptopis stage, which at an early instar (C) is characterized by the distinct development of the carapace and the elongation of the abdomen. The median eye is replaced by sessile rudiments of compound eyes con- cealed beneath the carapace. The appendages are those of the meta- nauplius. At a later calyptopis instar (D) the abdomen has become ( », 54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 segmented and the uropods are developed. In the furcillia stage (E, F) the larva begins to resemble the adult. The eyes are now stalked and project from beneath the carapace. The first furcillia instar (E) has still only the appendages of the metanauplius, but after the first moult the pereiopods appear as simple papillae, which later enlarge (F, G) and finally become biramous appendages. At the same time the pleopods are formed. According to Lebour (1925) in Nyctiphanes and Meganyctiphanes there are 12 furcillia instars sepa- rated by moults. In the final cyrtopia stage (G), after 8 to 13 moults according to the species, the young euphausiid acquires the adult structure with a complete set of appendages and luminescent organs. DECAPODA The decapod crustaceans include the shrimps, lobsters, crayfishes, and crabs. None of them exhibits any pronounced metamorphic changes during development or in the adult stage, but most of them go through stages of growth characterized chiefly by the successive development of sets of appendages. Only in the Penaeidae is there a free nauplius and a metanauplius. Most species hatch in a form .,called a zoea, in which the appendages following the second maxilli- ~~. peds are as yet undeveloped or are present as rudiments. With the functional completion of the pereiopods the larva is known as a mysis from its fancied resemblance to a member of the Mysidacea. Some species, however, go through the zoea stage in the egg and hatch as a mysis, and a few are almost completely developed in the adult form on leaving the egg. The decapod larvae are free swimming, and in general are fairly uniform in structure with a fully developed carapace and a long seg- mented abdomen. A few, however, take on unusual forms. Among the Sergestidae many of the larvae are characterized by a great de- velopment of long, often profusely branched spines on the thorax and abdomen. The rounded carapace of the palinuran Polycheles larva looks like a spiny burr, and others of the same group, known as phyllosome larvae, are broad, flat, and leaflike in shape. Presumably such forms are adaptations to buoyancy or floating. The Penaeidea.—In this order the family Penaeidae is of particular interest because it includes the only decapods that begin life as free- swimming nauplii. The fact that among the Malacostraca both the penaeids and the euphausiids hatch from the egg as nauplii may be taken as evidence that primarily all the crustaceans hatched at this early stage of development, and that later hatching among the higher Malacostraca is secondary, resulting from the earlier stages being NO. 10 CRUSTACEAN METAMORPHOSES—SNODGRASS 55 completed for better security in the egg. The life history of Penaeus setiferus (L.) is now well known from the studies of Pearson (1939) and Heegaard (1953), and will here be briefly reviewed from the papers by these two authors. The penaeid life history, moreover, will serve also as a good subject for a discussion of the significance of crustacean larval forms. Both the nauplius and the metanauplius of Penaeus (fig. 21 A, B) have long swimming appendages, but the alimentary canal is not yet developed and the larvae in these stages subsist on the yolk derived from the egg. In the metanauplius (B), however, the mandibles have acquired gnathal lobes on their bases, and rudiments of four pairs of postmandibular appendages are present, the last being those of the second maxillipeds (2M«pd). The metanauplius goes over into the zoeal stage, in which there are three instars. In the first zoea (C) the carapace has developed, the mandibles have become functional jaws, and the larva now takes its first external food. The following appendages have developed into biramous limbs, and the abdomen shows a faint trace of segmentation, but the larva apparently still swims by means of the antennae. In the third zoeal instar (D) the larva takes on something of the adult form (G). The carapace covers the thorax, and rudiments of the pereiopods (D, Prpds) are present, the abdomen is fully segmented but pleopods have not yet appeared, and the antennae are still the chief organs of propulsion. The third zoea is followed by the so-called mysis stage, which goes through two instars. In the first mysis (E) the pereiopods are all present and have long seta-bearing exopodites, which now assume the locomotor function, and the antennae are reduced. The abdomen has well-developed uropods, but pleopods are as yet absent. The next stage (fig. 21 F), known as the postmysis, or postlarva, more nearly resembles the adult. The pereiopods have lost their exopodites, and those of the first three pairs are chelate. Slender uniramous pleopods are present on the abdomen and are now the swimming organs as they are in the adult. In the adult (G) the pleopods have acquired the typical biramous structure, and a long filamentous flagellum arises from each second antenna. The life-history stages of Lucifer, as described by Brooks (1882), are similar to those of Penaeus, except that the larva hatches as a metanauplius and the animal takes on a different form in its preadult and mature stages. Numerous examples of the bizarre larvae of Sergestidae, characterized by long, branched spines on the thorax and abdomen, are illustrated by Gurney (1924). There has been much discussion among carcinologists as to whether or not the forms of decapod larvae have a phylogenetic significance. 56 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 Gurney (1942) has pointed out that “the larval stages of today pro- vide evidence for phylogeny, but indirectly,” since the ontogeny of an animal recapitulates the ontogeny of its ancestors. Fic. 21.—Decapoda: Penaeidea. Developmental stages of Penaeus setiferus (L.). (A-F from Pearson, 1939.) A, first nauplius. B, metanauplius. C, first protozoea. D, third protozoea. E, first mysis. F, first postmysis. G, adult. 2M xpd, second maxillipeds ; Prpds, rudiments of pereiopods. It is true that ontogenetic stages of a species may represent in a modified way adult ancestral stages of phylogenetic evolution. The adult ancestry of a crustacean, however, can go back only as far as NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 57 the primitive adult arthropod from which the Crustacea were evolved. Life-history stages representing adult crustacean ancestors, therefore, can be recapitulations only of forms that intervened in evolution between the primitive arthropod and the modern crustacean. On the assumption adopted in the early part of this paper as a basic concept, the primitive arthropod is presumed to have been an elongate, segmented animal with a pair of similar jointed appendages on each body segment (fig. 1C). From such a progenitor all the modern arthropods were evolved by special modifications, particularly of the appendages, according to the adopted way of living. Anaspides (D) may be taken as an example of a fairly generalized modern crustacean, but other crustaceans go through no developmental stage resembling Anaspides or any other form that might be intermediate between their adult structure and that of a primitive arthropod. The megalops of a crab undoubtedly represents an early crab form, but there is little evidence that the Crustacea in general recapitulate adult stages of crustacean ancestry or the adults of other species of lower rank in taxonomy. There is no reason to believe that the likeness of the “mysis” stage of the penaeid (fig. 21 E, F) to an adult Mysis is any- thing more than a superficial resemblance. Foxon (1936) has shown that the decapod larvae do not go through a typical euphausiid or mysid stage, and that neither the structure nor the function of the mysid appendages is recapitulated in other groups. The precocious development of the uropods before the pleopods are formed is ex- plained by Foxon (1934) as an adaptation to reverse movement. Most crustaceans develop by anamorphosis, but the anamorphic method of growth was established in the remote progenitors of the arthropods before the arthropods became arthropods. The embryo in the egg goes through the preanamorphic stages of its ancestors, and if it is hatched as a nauplius, the following ontogenetic stages recapitulate the anamorphic steps of precrustacean evolution. The larva, however, is destined to be a crustacean, it carries the genes of its species, and its crustacean destiny is thus stamped on it before it leaves the egg. Hence, from the beginning of its development the larva takes on crustacean characters, but the forms it assumes are ontogenetic and not recapitulations of adult crustacean evolution. When the larva is set free at a very immature stage it must be struc- turally adapted to the exigencies of an independent life, and it may be modified for a way of living that was not at all that of its ancestors. Thus the normal ontogenetic stages may take on metamorphic aberra- tions having no relation to anything in the past history of the animal 58 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I or to its own future adult stage. Such nonancestral forms are par- ticularly evident in parasitic species. Gurney (1942), referring to the progressive shift of the swimming function in the larva from the antennae to the pereiopods and finally to the pleopods, has expressed the idea that “the fundamental fact which determines the organization of the larva is the mode of locomo- tion.” However, it is to be presumed that the use of the pleopods for swimming was first established in the adult ancestors of such species. The nauplius naturally cannot swim in this ancestral manner, and must use what appendages it has. As the larva grows by the addition of segments and appendages it can more efficiently swim by making use of the pereiopods, and finally when the pleopods are developed it can swim in the adult manner. It is the progressive organization of the larva, therefore, that determines the mode of locomotion. The Macrura.—The macruran decapods are the lobsters and the crayfishes. The lobster, Homarus, according to S. I. Smith (1871- 1873) undergoes its early development in the egg and hatches at a stage when all the pereiopods are present and are equipped with feathery exopodites. This first free stage of the lobster (fig. 22 A), therefore, corresponds with the mysis stage of Penaeus (fig. 21 E). In the next instar the larva increases somewhat in size, and rudiments of pleopods appear on the abdomen. In the third instar (fig. 22 B) the young lobster attains a length of 12 to 13 mm. and much re- sembles the adult; the chelae are well developed, the pleopods are biramous, but the exopodites are still present on the pereiopods. Smith suggests that there is probably another instar intervening be- tween the third and the adult when the exopodites are lost, as in the postmysis of Penaeus (fig. 21 F). The fresh-water crayfishes, Astacus and Cambarus, hatch at a later stage of development than Homarus, when they have practically the adult structure except for the lack of the first and sixth pleopods. The Brachyura.—The brachyurans, or “short-tailed” decapods, are the ordinary crabs, so named because of the small size of the abdomen, which in the adult is carried bent forward beneath the thorax. The zoeal larvae are characterized in most species (fig. 23) by the presence of a long dorsal spine on the thorax and by the spinelike form of the rostrum, the two often projecting in a straight median line from opposite ends of the back. Some have also lateral spines. The larva swims with the large first and second maxillipeds, and the spines are supposed to assist in directing the course of the larva in the water or to help keep it afloat. The spines are absent in only a few species, as in the genus Ebalia and in members of the Pinnotheridae. The last NO. I0 CRUSTACEAN METAMORPHOSES SNODGRASS 59 zoea transforms into a preliminary crablike stage known as a megalops. The life history of the blue crab of the Chesapeake Bay, Callinectes sapidus Rathbun, has been studied by Churchill (1942), Hopkins (1944), and Sandoz and Hopkins (1944), and is typical of the de- velopment of most of the Brachyura. The young crab is sometimes hatched in a final embryonic stage called by Churchill a prezoea (fig. 23 A). It is still enclosed in a thin, transparent, closely fitting cuticle Fic. 22.—Decapoda: Macrura and Brachyura. Young stages. (A, B from S. I. Smith, 1871-1873; C from Cano, 1891.) A, Homarus americanus H. Milne Edw., first larval instar, zoea. B, same, third instar. C, Pilwmnus, a brachyuran crab, metazoea with partly developed chelipeds and pereiopods. that covers the spines, which will be exposed at the first moult. Sandoz and Hopkins say that emergence in the prezoeal stage results from unfavorable conditions at the time of hatching. The first free larva is a typical crab zoea (B) about 0.85 mm. in length. It has a short, rounded carapace and a long, slender, segmented abdomen. The last appendages are the large first and second maxillipeds, the exopodites of which are equipped with terminal fans of long featherlike bristles. The sixth segment of the abdomen is still united with the telson. In the second zoea (C) there is no essential change of structure, but the 60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 body and appendages have increased in size. Churchill describes five zoeal instars in Callinectes sapidus, but his figures of the third, fourth, and probably the fifth instar are said by Hopkins (1944) to be larvae of some other crab. The differences, however, are slight and pertain mostly to the number of setae on the appendages and spines on the abdomen. In the fifth instar (D) the last abdominal segment is sepa- rated from the telson and pleopods are present. About the only Fic. 23.—Decapoda: Brachyura. Larval stages of Callinectes sapidus Rathbun (from Churchill, 1942). A, prezoea. B, first zoea. C, second zoea. D, fifth zoea. E, megalops. metamorphic features of the crab zoea are the development of the dorsal and rostral spines and the adaptation of the maxillipeds for swimming. During the zoeal stage buds of the third maxillipeds and of the pereiopods appear on the thorax beneath the carapace and increase in length in successive instars, but they are not seen in Churchill’s figures (fig. 23). It seems hardly likely that the zoea shown at D of the figure could go over directly into the megalops (E). In the final zoea of other crabs, sometimes called a metazoea, the appendages be- hind the second maxillipeds are well developed, as shown by Cano NO. 10 CRUSTACEAN METAMORPHOSES—-SNODGRASS 61 (1891) in the metazoea of Pilumnus (fig. 22C). The first and sec- ond maxillipeds still have the zoeal structure, but they are followed by the third maxillipeds and five pairs of pereiopods, of which the first are strongly chelate. Moreover, all these newly developed ap- pendages except those of the last two pairs support branchial lobes on their bases. Similar advanced larval instars are shown for several other species of Brachyura by Lebour (1928). Hence, we should assume that there must be in Callinectes a metazoeal instar in which the thoracic appendages are in a state of development that could go over at one moult into the appendages of the megalops. In the life history of the crab there is no form corresponding to the mysis stage of Penaeus (fig. 21 E) or that of Homarus (fig. 22 A), but the meta- zoea might be likened to the postmysis of Penaeus. The megalops (fig. 23 E) is clearly a young crab, though it is only a few millimeters in length. The dorsal spine of the zoea has been shed with the larval cuticle (fig. 22 C) and the rostrum is shortened to the ordinary length. The swimming maxillipeds are transformed into feeding organs, and the other appendages are those of the adult. The prominent stalked eyes give the megalops its name (“bigeye”). An important feature of the megalops, however, is the extension of the abdomen from the thorax, which suggests that the megalops represents an adult ancestral form of the crab before the latter permanently flexed its abdomen forward beneath the thorax. The adult crab on issuing from the cuticle of the megalops is still a minute creature and goes through a large number of instars before becoming sexually mature, after which it may continue to moult at intervals. The habits of adult crabs are more various than those of the larvae. While most adult crabs live in the ocean and crawl on the bottom, some of them live in the shells of mollusks, in echinoderms, in cavities of corals, and in tubes of worms. Others have left the water for the land, where they dig deep burrows in the sand above high water, and still others go freely inland, even invading human habita- tions. The famous anomuran robber crab of the South Sea Islands is said to climb cocoanut trees for their nuts. Regardless of their habits or the nature of their dwellings, however, the brachyuran crabs have undergone little structural adaptation. They vary in size and shape, in the relative size of the chelae, and in the length of their legs, but in general they retain the typical crab structure. Among the Anomura, however, a pronounced adaptive modification of the body occurs in the hermit crabs that live in snail shells. The carapace of these crabs is weak and flexible. The abdomen is a long, soft sac that 62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 fills the cavity of the snail shell ; pleopods are present generally on the left side only, but the uropods are strong, recurved appendages evi- dently serving to secure the crab in its house. STOMATOPODA The stomatopods are an individualistic group of malacostracans having some relatively primitive features in combination with so many structural specializations that it is difficult to give them a definite place in taxonomy. The head of the adult animal (fig. 24 G), pro- jecting from beneath a small rostral lobe of the carapace, has a com- plex structure not found in any other crustacean. The short, narrow carapace covers only the gnathal region and the first four thoracic segments. The other four free segments of the thorax are symmetri- cal with the large abdomen, and appear to be a part of it except for the leglike appendages borne on the last three. The limbs of the first, third, fourth, and fifth thoracic segments are turned forward and each bears a small apical chela; but those of the second segment (2L) are huge raptorial organs in which the terminal segments are long, strongly toothed claws, each closing tightly against the penultimate segment, giving the stomatopod its likeness to the insect praying mantis (which is not responsible for its name). The large abdomen has five pairs of pleopods, and the stomatopod gills are borne on the pleopods. The uropods are large, biramous appendages; the telson is a broad spiny plate. The adult stomatopods are mostly littoral in their habits. Though they swim freely, they live principally in burrows in the sand or mud of the bottom. The females lay their eggs in a mass beneath the fore part of the body, where they are held between the raptorial legs by the four small chelate legs of the thorax. The eggs are carried in this manner until the young larvae emerge, a period said by Giesbrecht (1910) to last for Io or 11 weeks. The young stomatopods are hatched in two different larval forms, which seem to have no developmental relation to each other. Our best source of information on the larval stages will be Giesbrecht’s (1910) elaborate monograph on Mediterranean species. Gurney (1946) gives descriptions and good illustrations of various stomato- pod larvae, but no full account of the life history of any one species. Alikunhi (1952) describes and figures particularly the last-stage larvae of Indian species. The simpler first-stage larval form pertains to species of Lysio- squilla and Coromida, and is termed by Giesbrecht an antezoea, This NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 63 larva (fig. 24 A) is from 2 to 2.25 mm. in length. The thorax is fully segmented and is entirely covered by the carapace. The abdomen (AbD) is either unsegmented and entirely united with the telson (Tel) aig TSN 2Ant-f LEGA y / ff a {i s TS, pe 2L - Fic. 24.—Stomatopoda. Larval stages and an adult. (A-E from Giesbrecht, 1910; F from Alikunhi, 1952.) A, an antezoea larva. B, Squilla mantis Latr., first propelagic stage. C, same, second propelagic stage. D, same, first pelagic stage, dorsal. E, same, first pelagic stage, lateral. F, Squilla latreillei, last pelagic larval stage. G. Squilla mantis, adult male. Ab, abdomen; 1Ant, first antenna; 2Ant, second antenna; E, eye; 1L,2L,5L, 8L, first, second, fifth, and eighth thoracic appendages; Tel, telson. in a wide, fan-shaped plate, or one or two anterior segments may be free. The eyes (£) are large but sessile. The first five thoracic seg- ments bear each a pair of small, biramous appendages (7L, 5L) used for swimming. The antezoeal larva is pelagic. During subsequent 64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I stages of its growth, the abdominal segments are successively sepa- rated from the telson and acquire pleopods. The five thoracic ap- pendages lose their exopodites and take on the adult form, those of the second segment becoming typical raptorial fangs in the fourth instar. Later the appendages appear on the last three segments of the thorax. In the second instar the eyes are stalked. The other first-larval form, termed a pseudozoea (fig. 24 B), occurs in species of Squilla, Gonodactylus, and probably of other genera. The eyes in this form are stalked at hatching. The thorax is fully segmented, but only the first two segments bear appendages, and those of the second segment are raptorial fangs. The short carapace has small spines on its anterior and posterior angles, and leaves four posterior thoracic segments uncovered. The abdomen has five free segments, of which the first four bear pleopods, but the sixth is still united with the telson. Squilla mantis, according to Giesbrecht, goes through 10 larval instars. The first two live on the bottom, but after the second instar the larva becomes pelagic, swimming with the pleopods. In the sec- ond propelagic instar (fig. 24 C) there is little change from the first (B) except for an increase in size and a lengthening of the posterior carapace spines, which in the first pelagic instar (D, E) become much longer and widely divergent. In the third pelagic instar rudiments of the third, fourth, and fifth thoracic appendages appear, and become longer in the next stage, when also the appendages of segments six, seven, and eight are developed. The sixth segment of the abdomen becomes free from the telson in the seventh instar. The last pelagic larva (F) has essentially the structure of the adult (G) except for the large carapace, which now covers all but one of the thoracic segments. After about the fifth instar, Giesbrecht says, the two larval forms, originating with the antezoea and the pseudozoea, become struc- turally alike. The principal structural changes during the life of the stomatopod take place at the transformation of the larva (fig. 24 F) into the adult (G). Even here, however, the only essential change affects the carapace, which is much shortened and narrowed and loses its pos- terior spines. Instead of covering most of the thorax as in the larva (D, F) the carapace of the adult leaves the last four thoracic seg- ments exposed. In this respect the carapace reverts to its relative length in the first propelagic larva (B). It is evident, therefore, that the larval development of the back shield is a metamorphic adaptation to the pelagic life of the larva, probably to assist in keeping the larva afloat. The relative length of the larval carapace varies in different NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 65 species. In some forms the last four thoracic segments are not cov- ered, as in the adult of Squilla (G), in others such as Squilla latreillet (F) only the eighth segment is exposed in the larva, while in species of Lysiosquilla the carapace of the last larva may cover the entire thorax and the first two abdominal segments. Probably these varia- tions in the length of the larval carapace are only differences in the extent to which the free posterior margin is produced beyond the attachment of the plate on the third or fourth segment of the adult thorax. Otherwise the changes during the growth of the larva are ‘merely developmental stages of growth and have no metamorphic value. It is difficult even to see any functional reason for the differ- ences between the two larval forms on hatching. IV. STRUCTURE AND EVOLUTION OF ARTHROPOD APPENDAGES Inasmuch as changes in the form and function of the appendages are important features in the metamorphoses of Crustacea, and various conflicting views have been held concerning the primitive nature and the evolution of arthropod limbs, we must give some attention to this controversial subject. Most studies on the comparative structure of the arthropod ap- pendages, and deductions as to the origin and primitive form of the limbs give the impression that conclusions have resulted too much from an attempt to fit the facts into a preconceived theory. Widely accepted has been the idea that the primitive appendage was a biramous limb; and many carcinologists would derive all kinds of arthropod appendages from an original phyllopodial type of limb, such as that of the branchiopod crustaceans. The trilobites are among the oldest known arthropods, and, with respect to their appendages, they are the most generalized, since all the postoral limbs are fully segmented legs. The base of each leg bears a branched lateral process (fig. 25 A, Eppd), which, arising on the coxa, is clearly an epipodite and hence cannot be an equivalent of the crustacean exopodite, which by definition is an exite of the basipodite. The trilobite limb, therefore, is not “biramous” in the manner of a crustacean limb, and hence does not relate the trilobites to the Crustacea. Raymond (1920), however, explicitly states the opposite view. “The trilobites,” he says, “are themselves crustaceans, as is amply proven by their biramous appendages.” More recently, Hee- gaard (1947) has argued that the trilobite limb is truly biramous, in spite of the evident coxal position of the “exopodite,” and he further attempts to show that remnants of a primitive biramous structure are 66 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 to be found in various modern arthropods other than the Crustacea. His discussion, however, takes too many liberties with simple visible facts in an endeavor to fit them into a consistent scheme of structure. The studies of trilobite appendages by Stormer (1939) leave little doubt that the trilobite leg (fig. 25 A) is simply a uniramous, seg- mented limb with a coxal epipodite that was perhaps a gill. Stormer’s contention, however, that a narrow ring at the base of the coxa is a Fic. 25.—Examples of segmentation of arthropod legs. A, leg of a trilobite (from Stérmer, 1939). B, leg of Marella (adapted from Walcott, 1931). C, leg of Burgessia (from Walcott, 1931). D, leg of solpugid arachnid. E, leg of a chilopod, Lithobius. F, leg of a decapod, Cambarus. Crppd, carpopodite (tibia) ; Capd, coxopodite (coxa) ; Dactpd, dactylopodite (pretarsus) ; Eppd, epipodite; Mrpd, meropodite (femur) ; Pat, patella; Propd, propodite (tarsus). precoxal segment is questionable. The coxa of other arthropods is often marked by a circular groove near the base that forms an internal strengthening ridge giving attachment to the body muscles of the limb. In the trilobite leg the large coxopodite should be the movable basal segment of the limb and not the narrow “precoxa.” The idea that the primitive arthropod limb was a flat, lobulated appendage of the phyllopodial type has been accepted by some carcin- ologists regardless of the fact that the limbs of the trilobites (fig. 25 A) and of associated fossil forms such as Marella (B) and Burgessia (C) are slender jointed legs, as are those of nearly all modern arthropods (D, E, F), including the Malacostraca (F). NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 67 Walcott (1931), for example, in discussing his Burgess Shale fossils seems to accept this theory without question when he says: “The biramous limb of Marella, like that of the trilobite, undoubtedly passed through the foliaceous or multiramous limb stage in its evolu- tion, probably in pre-Cambrian time.” There is no disproving this idea, which should apply to the other arthropods as well, but such implicit faith in a theory is hard to understand. On the other hand, Raymond (1920) says the theory of the phyllo- pod origin of the arthropod limb “has been completely upset” by the finding of such “undoubted branchiopods” as Burgessia in the Middle Cambrian with trilobitelike legs. The same idea has been expressed by Heegaard (1947) in his statement that the “undoubted branchio- pods” found by Walcott in the Middle Cambrian having trilobite legs show that “it can no longer be held that the phyllopodial limbs are primitive.” The writer fully agrees with this conclusion, but for different reasons than those given by Raymond and Heegaard. Such fossils as Burgessia and Marella are certainly not “undoubted” branchiopods. Walcott (1931) says of Marella that it is a less primi- tive form than the Apodidae and more primitive than the trilobites, but is nearer to the latter than to the former. Among the Middle Cambrian fossils, however, is a form, Opabina regalis Walcott, par- ticularly studied by Hutchinson (1930), which evidently 7s an anostra- can branchiopod with foliaceous appendages. Another popular belief concerning the derivation of the arthropod limb, taken to support the theory of its biramous phyllopodial origin, is that the limb was evolved from the polychaete parapodium. Reasons have already been given in section I of this paper for believing that the annelids have only a remote connection with the arthropod pro- genitors. Certainly the arthropods can have no relation to modern polychaetes, which are highly specialized annelids and could give rise only to more polychaetes. The appendages of the worm, though they are bilobed flaps, have a lateral position on the body (fig. 26 A), and there is nothing in their structure having any likeness to an arthropod limb at any stage of its development. The parapodium bears two bundles of bristles supported on a pair of long internal rods giving attachment to muscles. Its only common feature with an arthropod limb is that, being a locomotor organ, it is movable forward and back- ward by body muscles. In short, the idea that.the arthropod append- ages were derived from annelid parapodia appears to be just another case of excessive zeal for generalization. Among modern wormlike animals those closest to the arthropods are the Onychophora; some zoologists have even included the ony- 68 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 chophorans in the Arthropoda. Though a modern onychophoran shows no external segmentation of the body in the adult stage, the segmented repetition of internal organs and the complete body seg- mentation of the embryo leave no doubt that the Onychophora are AlCnl DV \ / B Fic. 26.—Diagrammatic transverse sections of Nereus (A) and Peripatus (B), showing comparative structure of the appendages of a polychaete annelid and an onychophoran. AlCnl, alimentary canal; Com, nerve commissure; DDph, dorsal diaphragm ; dm, dorsal muscles; DS, dorsal sinus; Dsp, dissepiment; DV, dorsal blood vessel; dum, dorsoventral muscles; NC, lateral nerve cord; Nph, nephridium ; Npr, nephropore; Papd, parapodium; S/mGld, slime gland; vm, ventral muscles ; VNC, ventral nerve cord. fundamentally metameric animals. The body cavity is undivided by dessepiments, the primitive coelom is represented only by the lumina of the nephridia and the gonads, and the embryogeny of the Onycho- phora gives the key to the early embryonic development of the arthro- pods. The onychophoran legs have a lateroventral position on the body (fig. 26B) as in the arthropods, in contrast to the lateral posi- NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 69 tion of the polychaete parapodia (A). The nephridia (B, Nph) and the primitive genital ducts open mesad of the leg bases suggestive of their openings on the coxae in many of the arthropods. Though modern Onychophora are terrestrial animals, there can be little doubt that they had aquatic ancestral relatives represented by the Cambrian Aysheaia of the Burgess Shale, and perhaps by the Pre-Cambrian Xenusion described by Heymons (1928). The arthropod limbs are developed on the embryo from latero- ventral budlike rudiments that lengthen and become segmented. We may therefore suppose that from the ancestral onychophorans (fig. 1 A) a form was evolved with longer legs (B), which later, with sclerotization of the integument, became the jointed appendages of the ancestral arthropods (C). It then required a long period of Pre- Cambrian evolution to produce a trilobite on the one hand, and some ancestral form of crustacean on the other. The differentiation between the two groups, however, was first in the form of the body, not in that of the appendages, as seen in the legs of a trilobite (fig. 25 A) and those of Marella and Burgessia (B, C). Though there is no valid reason for regarding the primitive arthropod appendage as being a biramous limb, the crustacean appendages later acquired their charac- teristic biramous structure, which is usually lost in the ambulatory limbs (F). Many carcinologists hold the view that the phyllopodial type of limb is primitive, at least for the Crustacea, and this concept has been well elucidated by Borradaile (1926a, 1926b). It is supposed that the primitive crustacean appendage was a flat, unsclerotized lobe with a fringe of hairs on the mesal border. Then the inner margin was broken up by the development of a series of endites. Next, the limb became more rigid by a sclerotization of the integument, but this necessitated lines of flexibility that led to a system of jointing, and naturally the joints were formed between the endites. Thus the endites are explained as the precursors of the later developed limb segments. Finally, with the departure from the phyllopodial form and the suppression of the endites, some of the limbs became slender, seg- mented, leglike appendages. In favor of this theory it may be noted that in many of the branchiopod appendages there are six endites on the mesal margin and a free lobe at the apex (fig. 27 A,B). If all the parts of such a limb became segments there would be seven seg- ments in all, the terminal lobe becoming the dactylopodite, which gives the usual number of limb segments in the Crustacea generally, though Borradaile holds that the maximum number is nine, which would include the doubtful “precoxa” of the trilobite. 7O SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 eZ | | — OS Fic. 27,—Examples of branchiopod appendages. A, Branchipus stagnalis, Anostraca, thoracic limb (from Claus, 1873). B, Branchipus serratus, male, eighth thoracic limb. C, Apus longicaudata, Noto- straca, second maxilliped. D, same, first maxilliped. E, same, thoracic limb from middle of body. F, Daphnia magna, Cladocera, third thoracic appendage (from Hansen, 1925). G, Estheria clarkii, Conchostraca, second thoracic limb. H, same, terminal segment. I, same, left limb from middle of body. Bspd, basipodite; Crppd, carpopodite; Capd, coxopodite; Dactpd, dactylopo- dite; Iscpd, ischiopodite; Mrpd, meropodite; Propd, propodite; 1-6, endites. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS 7a There are two chief objections to this phyllopod theory of the origin of jointed crustacean limbs. First, it gives no explanation of the origin of the similarly jointed legs of other arthropods, except by the wholly unsupported assumption that they likewise were developed from phyllopodial limbs. Second, the ontogenetic development of the crusta- cean appendages themselves gives no evidence of a phyllopodial origin, and suggests, on the contrary, that the phyllopodium has been evolved from an ambulatory leg. The study by Heath (1924) of the postembryonic development of the branchiopod Branchinecta occidentalis shows very clearly that the limbs arise as simple, lateroventral lobes of the body segments (fig. 3B). Instead of taking on a phyllopodial shape, the rudiments grow out first in a slender leglike form (C, D). On the inner margins of the appendages at this stage there are indentations suggestive of an incipient segmentation, and at the apex is a terminal lobe. Only at a later stage (E) do the appendages become broad overlapping flaps. Finally in the adult (F) the appendages have taken on the form of typical unsegmented phyllopodia with three large flat exites, six endites, and a free, independently musculated terminal lobe. Clearly, these appendages in their ontogenetic development undergo a meta- morphosis from an ambulatory leg into a phyllopodium. Though Heath himself did not have this phase of the subject in mind, his pictures speak for themselves. Conversely, as seen in Heegaard’s (1953) account of the post- embryonic stages of the decapod crustacean Penaeus setiferus, the rudiments of the pereiopods develop directly into legs without under- going any stage suggestive of a phyllopodial origin. The pereiopods appear during the second protozoeal stage as simple lobes on their re- spective body segments (fig. 28 A). In the third protozoea they take on a biramous structure (B), in which the protopodite, at first un- divided, bears a short unsegmented endopodite and a longer exopodite. In the second mysis stage (C) the limbs attain a fully segmented struc- ture by the division of the protopodite into two segments and the endopodite into five, with a terminal chela on each of the first three. The exopodites are now large seta-bearing branches of the basipodites used for swimming. In the postmysis (D) the pereiopods have become essentially uniramous by the reduction of the exopodites to small lobes, and the swimming function has been taken over by the pleopods. This condition is retained in the adult. If the pereiopods of Penaeus had a phyllopodial origin in their phylogeny, there is nothing to suggest it in their ontogeny. The mouth-part appendages proceed along their own lines of development to serve the special functions they have assumed 72 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 as organs of feeding. The pleopods (E) and the uropods (I), on the other hand, appear to remain in an early stage of development repre- sented by the simple, unsegmented biramous stage of the pereio- pods (B). The swimming appendages of the anostracan branchiopods so regu- larly have six mesal lobes (figs. 3 F, 27 A, B) and a movable terminal Fic. 28.—Development of the pereiopods and pleopods of Penaeus setiferus (L.) (from Heegaard, 1953). A, rudiments of pereiopods on thoracic segments of second protozoea. B, pereiopod of third protozoea. C, pereiopod of second mysis instar. D, third pereiopod of postmysis. E, second pleopod of young adult. F, uropod of post- mysis. Bspd, basipodite; Cxpd, coxopodite; Endpd, endopodite; Expd, exopodite; Prpds, pereiopod rudiments ; Prtpd, protopodite. lobe as to suggest that the six endites represent the first six segments of a leg (coxopodite to propodite) and the independently musculated apical lobe the dactylopodite. Yet, the sixth endite is commonly inter- preted as the endopodite and the apical lobe as the exopodite. In the notostracan Apus, however, the second maxilliped (fig. 27 C) is a seven-segmented leg ending with a clawlike dactylopodite (Dactpd) and having an endite on each of the other segments except the ischiopodite. The first maxilliped of Apus (D) is somewhat simpli- NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 73 fied, but the swimming appendages (E) clearly retain the structure of the second maxilliped. In other branchiopods the appendages may be variously reduced (F, G, H, 1) obscuring the basic leg structure. The segmentation of the arthropod legs is surprisingly constant ; variations result from the elimination of segments, seldom from addi- tion, though the propodite (tarsus) is generally rendered flexible by subdivision. If all the podomeres in the legs of the trilobite (fig. 25A) and Marella (B) are true musculated segments, the ancient arthropods had eight limb segments, including a small apical dactylopo- dite, or pretarsus, and thus possessed all the segments that are present in any of the legs of modern arthropods. Among the latter, eight segments are present in the Pycnogonida and in some of the legs of the arachnid Solpugidae (D), but in most of the arachnids the leg has only seven segments by the elimination of the third segment from the base. The segment beyond the knee bend (D, Pat), which is the fifth segment in the trilobite leg (A), is called the patella, though it might appear to correspond with the carpopodite (tibia) in the leg of a centipede (E) or a decapod (F). Yet there are three segments beyond it in the spider leg, and only two in the other arthropods. In the latter, therefore, either two original segments in the distal part of the leg are united, or one has been eliminated. The legs of the chilopods and the decapods (E, F) have seven segments; the insect leg has only six segments because of the apparent union of the ischiopodite (second trochanter, or prefemur) with the meropodite (femur). Though the primitive arthropods (fig. 1C) undoubtedly were aquatic, they were walking animals provided with jointed limbs, and probably lived on plants in shallow water near the shore. Their habits may have been similar to those of the modern Anaspides (D). The typical jointed ambulatory leg has been retained in all modern arthro- pods, except in those crustaceans in which it has been modified for swimming, but even the phyllopodium preserves evidence of the seven- segmented structure of a walking leg. It would appear that the primi- tive arthropods had more legs than they needed for walking, and because of this fact their descendants have been able to reconstruct many of them into the great variety of appendicular organs possessed by modern forms. The arthropods owe what they are, as well as their name, to their jointed appendages. REFERENCES ALIKUNHI, K. H. 1952. An account of the stomatopod larvae of the Madras plankton. Rec. Indian Mus., vol. 49, pp. 239-310, 25 figs. 74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I Buiss, Dororuy E., and WEtsH, J. H. 1952. The neurosecretory system of brachyuran Crustacea. Biol. Bull. vol. 103, pp. 157-169, 9 figs. BorrRADAILE, L. A. 1926a. Notes upon crustacean limbs. Ann. Mag. Nat. Hist., ser. 9, vol. 17, Pp. 193-213, 5 pls. 1926b. On the primitive phyllopodium. Ann. Mag. Nat. Hist., ser. 9, vol. 18, pp. 16-18. Brooks, W. K. 1882. Lucifer: A study in morphology. Philos. Trans. Roy. Soc. London, vol. 173, pp. 57-137, I1 pls. Cannon, H. G. 1926. On the feeding mechanism of a freshwater ostracod, Pionocypris vidua (O. F. Miller). Journ. Linn. Soc. London, Zool., vol. 36, pp. 325- 335, 5 figs., 1 pl. CANo, G. 1891. Sviluppo postembrionale dei Cancridae. Bull. Soc. Ent. Italiana, vol. 23, pp. 146-158, 2 pls. Capart, A. 1948. Le Lernaeocera branchialis. La Cellule, vol. 52, pp. 159-212, 22 text figs., 7 pls. CAULLERY, M. 1908. Recherches sur les Liriopsidae, Epicarides cryptonisciens parasites des Rhizocéphales. Mitt. Zool. Stat. Neapel, vol. 18, pp. 583-643, 8 figs., pl. 26. CHURCHILL, E. P. 1942. The zoeal stages of the blue crab, Callinectes sapidus Rathbun, Chesa- peake Biol. Lab., Publ. No. 49, 25 pp., 4 pls. CLaus, C. 1863. Die freilebenden Copepoden. 230 pp., 37 pls. Leipzig. 1873. Kenntnis des Baues und der Entwicklung von Branchipus stagnalis und Apus cancriformis. Abh. Konig. Ges. Wiss. Gottingen, vol. 18, pp. 1-48, 8 pls. DELAGE, Y. 1884. Evolution de la Sacculine (Sacculina carcini Thomps.). Arch. Zool. Exp. Gén., sér. 2, vol. 2, pp. 417-736, pls. 22-30. DretricH, W. 1915. Die Metamorphose der freilebenden Siisswasserkopepoden. I. Zeit- schr. wiss. Zool., vol. 113, pp. 252-324, 19 figs. DoocuHin, H. D. 1951. The morphology of Balanus improvisus Darwin and Balanus amphi- trite niveus Darwin during initial attachment and metamorphosis. Bull. Marine Sci. Gulf and Caribbean, vol. 1, pp. 15-30, 7 figs. EcHALIER, G. 1954. Recherches expérimentales sur le rdle de “l’organe Y” dans la mue de Carcinus moenas (L.), Crustacé Décapode. Compt. Rend. Acad. Sci. Paris, vol. 238, pp. 523-525. FisHeEr, W. K. to11. Asteroidea of the North Pacific and adjacent waters. U.S. Nat. Mus. Bull. 76, Pt. 1, 419 pp., 122 pls. NO. IO CRUSTACEAN METAMORPHOSES—SNODGRASS | 75 Foxon, G. E. H. 1934. Notes on the swimming methods and habits of certain crustacean larvae. Journ. Marine Biol. Assoc. United Kingdom, vol. 19, pp. 829-850, 7 figs. 1936. A note on recapitulation in the larvae of the decapod Crustacea. Ann. Mag. Nat. Hist., ser. 10, vol. 18, pp. 117-123, 4 figs. FRANSEMEIER, LUISE. 1939. Zur Frage der Herkunft des metanauplialen Mesoderms und die Seg- mentbildung bei Artemia salina Leach. Zeitschr. wiss. Zool., vol. 152, pp. 439-472, 17 figs. Gabe, M. 1953. Sur l’existence, chez quelques Crustacés Malacostracés, d’un organe comparable a la glande de la mue des Insectes. Compt. Rend. Acad. Sci. Paris, vol. 237, No. 18, pp. 1111-1113. GIESBRECHT, W. 1910. Stomatopoden, Ist part. Fauna Flora Golfes Neapel, Monogr. 33, 239 pp., 12 figs., 11 pls. Gurney, R. 1924. Decapod larvae. British Antarctic (Terra Nova) Expedition, 1910. Zool., vol. 8, No. 2, pp. 37-202, 78 figs. 1926. Life-history problems of crustacean larvae. Discovery Reports, vol. 7, PP. 350-353, 7 figs. 1942. Larvae of decapod Crustacea. 306 pp., 122 figs. Ray Society, London. 1946. Notes on stomatopod larvae. Proc. Zool. Soc. London, vol. 116, pp. 133-175, 14 figs. Hansen, J. H. 1925. Studies on Arthropoda. II. Crustacea. 176 pp., 8 pls. Copenhagen. Heatu, H. 1924. The external development of certain phyllopods. Journ. Morph., vol. 38, pp. 453-483, 5 pls. HEEGAARD, P. 1947. Contribution to the phylogeny of the arthropods. Copepoda. Skrif. Univ. Zool. Mus., Kgbenhavn, vol. 8, 236 pp., 27 pls. 1948. Larval stages of Meganyctiphanes (Euphausiacea) and some general phylogenetic remarks. Medd. Komm. Danmarks Fiskeri og Havun- ders6gelser, ser. Plankton, vol. 5, No. 1, 27 pp., 4 pls. 1951. Antarctic parasitic copepods and an ascothoracid cirriped from brittle- stars. Vid. Medd. Dansk Naturh. Foren., vol. 113, pp. 171-190, 7 figs., 2 pls. 1953. Observations on spawning and larval history of the shrimp, Penaeus setiferus (L.). Publ. Inst. Marine Sci., vol. 3, pp. 74-105, 133 figs. Heymons, R. 1928. Uber Morphologie und verwandtschaftliche Beziehungen des Xenusion auerswalde Pomp. aus dem Algonkium, Zeitschr. Morph. und Okol. Tiere, vol. 10, pp. 307-329, 7 figs. Hinton, H. E. 1946. Concealed phases in the metamorphosis of insects. Nature, vol. 157, PP. 552-553, I fig. 76 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I Hopkins, S. H. 1944. The external morphology of the third and fourth zoeal stages of the blue crab, Callinectes sapidus Rathbun. Biol. Bull., vol. 87, pp. 145- 152, 2 pls. Hutcuinson, G. E. 1930. Restudy of some Burgess Shale fossils. Proc. U.S. Nat. Mus., vol. 78, art. II, pp. I-24, 5 figs., I pl. Iwanorr, P. P. 1928. Die Entwicklung der Larvalsegmente bei den Anneliden. Zeitschr. Morph. und Okol. Tiere, vol. 10, pp. 62-161, 40 figs., 7 pls. KEsLING, R. V. 1951. The morphology of ostracod molt stages. Illinois Biol. Monogr., vol. 21, 324 pp., 34 figs., 96 pls. KwnipowItTscu, N. 1890. Dendrogaster astericola nov. g. et sp., eine neue Form aus der Gruppe Ascothoracica. Biol. Centralb., vol. 10, pp. 707-711, 3 figs. KNow Les, F. G. W. 1953. Endocrine activity in the crustacean nervous system. Proc. Roy. Soc. London, ser. B, vol. 141, pp. 248-267, 8 figs., 3 pls. Lespour, Marie V. 1916. Stages in the life history of Calanus finmarchicus (Gunnerus), experi- mentally reared by Mr. L. R. Crawshay in the Plymouth laboratory. Journ. Marine Biol. Assoc. United Kingdom, vol. 11, pp. 1-17, 5 pls. 1925. The Euphausiidae in the neighborhood of Plymouth. II. Nyctiphanes couchi and Meganyctiphanes norvegica. Journ. Marine Biol. Assoc. United Kingdom, vol. 8, pp. 810-829, 9 pls. 1928. The larval stages of the Plymouth Brachyura. Proc. Zool. Soc. London, 1928, pp. 473-560, 5 figs., 16 pls. Mataguin, A. 1901. Le parasitisme €évolutif des Monstrillides (Crustacés Copépodes). Arch. Zool. Exp. Gén., sér. 3, vol. 9, pp. 81-232, 8 figs., 7 pls. Manton, S. M. 1928. On the embryology of a mysid crustacean, Hemimysis lamornae. Philos. Trans. Roy. Soc. London, ser. B, vol. 216, pp. 363-463, 32 figs., 5 pls. 1934. On the embryology of the crustacean Nebalia bipes. Philos. Trans. Roy. Soc. London, ser. B, vol. 223, pp. 163-238, 17 figs., 7 pls. Monon, T. 1926. Les Gnathiidae. Mém. Soc. Sci. Nat. Maroc, vol. 13, 668 pp., 275 figs., I pl. Norman, A. M. 1913. Synagoga mira, a crustacean of the order Ascothoracica. Trans. Linn. Soc. London, Zool., vol. 11, pp. 161-166, 3 pls. OEHMICHEN, A. 1921. Die Entwicklung der dusseren Form des Branchipus grubei Dyb. Zool. Anz., vol. 53, pp. 241-253, 18 figs. Oxana, Y. K. 1938. Les cirripédes ascothoraciques. Trav. Stat. Zool. Wimereux, vol. 13, pp. 489-512, 18 figs., 1 pl. NO. I0 CRUSTACEAN METAMORPHOSES—SNODGRASS 77 Ovip (Pustius Oviprus Naso). A.D. 7. Metamorphoses. PANOUSE, J. B. 1946. Recherches sur les phénoménes humoraux chez les Crustacés. Ann. Inst. Océanographique, vol. 23, pp. 65-147, 49 figs. Fassano, L. M. 1953. Neurosecretory control of molting in crabs by the X-organ sinus gland complex. Physiol. Comp. et Oecol., vol. 3, pp. 155-189, 6 figs. PEARSON, J. C. 1939. The early life histories of some American Penaeidae, chiefly the com- mercial shrimp Penaeus setiferus (Linn.). U.S. Bur. Fisheries Bull. 30, 73 pp., 67 figs. PEDASCHENKO, D. 1898. Embryonalentwicklung und Metamorphose von Lernaea branchialis L. Trav. Soc. Imp. Nat. St. Pétersbourg (Trudy), vol. 26, livr. 4, No. 7, 307 pp., 6 pls., German text, pp. 247-307. Ports, F. A. 1915. On the rhizocephalan genus Thompsonia and its relation to the evolu- tion of the group. Pap. Dept. Marine Biol. Carnegie Inst. Wash- ington, vol. 8, pp. 1-32, 12 figs., 2 pls. Raymonp, P. E. 1920. Phylogeny of the Arthropoda with especial references to the trilobites. Amer. Nat., vol. 54, pp. 398-413. RUNNSTROM, S. 1924-1925. Zur Biologie und Entwicklung von Balanus blanoides (Linné). Bergens Mus. Aarb., 1924-25, No. 5, 46 pp., 22 figs. Sanpoz, Mitprep, and Hopkins, S. H. 1944. Zoeal larvae of the blue crab Callinectes sapidus Rathbun. Journ. Washington Acad. Sci., vol. 34, pp. 132-133. SCHEERER-OSTERMEYER, ELSE. 1940. Beitrag zur Entwicklungsgeschichte der Siisswasserostrakoden. Zool. Jahrb., Anat., vol. 66, pp. 349-370, 5 figs. SCHREIBER, ERNA. 1922. Beitrage zur Kenntnis der Morphologie, Entwicklung und Lebens- weise der Siisswasser-Ostracoden. Zool. Jahrb., Anat., vol. 43, pp. 485-538, 24 figs. SCHUURMANS-STEKHOVEN, J. H. 1930. Beobachtungen zur Morphologie und Physiologie der Lernaeocera branchialis L., und der Lernaeocera lusci Bassett-Smith (Crustacea- Parasitica). Zeitschr. Parasitenkunde, vol. 8, pp. 659-6096, 28 figs. Scott, A. 1901. On the fish parasites, Lepeophtheirus and Lernaea. Trans. Liverpool Biol. Soc., vol. 15, pp. 188-241, 5 pls. SmiTH, G. 1906. Rhizocephala. Fauna Flora Golfes Neapel, Monogr. 29, 123 pp., 24 figs., 8 pls. SmiTH, S. I. 1871-1873. The early stages of the American lobster (Homarus americanus Edwards). Trans. Connecticut Acad. Arts and Sci., vol. 2, pp. 351- 381, 5 pls. 78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 131 Snoperass, R. E. 1938. Evolution of the Annelida, Onychophora, and Arthropoda. Smith- sonian Misc. Coll., vol. 97, No. 6, 159 pp., 54 figs. SoLLaup, E. 1923. Recherches sur l’embryogénie des Crustacés décapodes de la sous- famille des ‘““Palemoninae.” Bull. Biol. France et Belgique, Suppl. 5, 234 pp. 17 figs., 5 pls. Sproston, N. 1942. The developmental stages of Lernaeocera branchialis (Linn.). Journ. Marine Biol. Assoc. United Kingdom, vol. 25, pp. 441-466, 5 figs. STORMER, L. 1939. Studies on trilobite morphology. I. The thoracic appendages and their phylogenetic significance. Norsk Geol. Tidsskr., vol. 19, pp. 143-273, 35 figs., 12 pls. VEILLET, A. 1945. Recherches sur le parasitisme des crabes et des galathées par les Rhizocéphales et les Epicarides. Ann. Inst. Océanographique, vol. 22, pp. 193-341, 31 figs. Watcott, C. D. 1931. Addenda to description of Burgess Shale fossils. (With notes by C. E. Resser.) Smithsonian Misc. Coll., vol. 85, No. 3, 46 pp., 11 figs., 23 pls. WIitson, C. B. 1905-1907. North American parasitic copepods belonging to the family Caligidae. Pt. 1, Proc. U.S. Nat. Mus., vol. 28, pp. 479-672, 51 figs., 25 pls.; pt. 2, vol. 31, pp. 669-720, 19 figs., 6 pls.; pts. 3 and 4, vol. 33, pp. 223-490, 18 figs., 27 pls. 1911. North American parasitic copepods. Pt. 9. The Lernaeopodidae. Proc. U.S. Nat. Mus., vol. 39, pp. 189-226, 18 pls. 1915. North American parasitic copepods belonging to the Lernaeopodidae, with a revision of the entire family. Proc. U.S. Nat. Mus., vol. 47, pp. 565-720, 15 figs., 32 pls. 1917. North American parasitic copepods belonging to the Lernaeidae, with a revision of the entire family. Proc. U.S. Nat. Mus., vol. 53, pp. 1-150, 4 figs., 21 pls. 192I1a. New species and a new genus of parasitic copepods. Proc. U.S. Nat. Mus., vol. 50, pp. I-17, 7 pls. 1921b. The North American semiparasitic copepods of the genus Clausidium. Proc. U.S. Nat. Mus., vol. 59, pp. 425-431, 2 pls. ZIEGELMAYER, W. 1925. Metamorphose und Wachstum der Cyclopiden. Zeitschr. wiss. Zool., vol. 126, pp. 493-570, 20 figs., 2 pls. UL 5 a ory Bhi i. a if ; gh, 9 Ae re A, 4d. : } : j re i Re ? “Stris. Mii fit GLE Poe eiie ioc a Aa | ee, Ona Ri et . oe. i) ET Cs be : 7 Pha) a hyip a aM th Wie; ea, 4 ee ygety-Le vY Aa : . can (: i Et a an s i fi : ire a can Le ety 7 ir, i oD ic a er i sie be a a as my me y 7 ; wi i, ar T one r 1) Ub ores IPRA) ah a a ie he vn i Mi me 1 oF Mi nie! Le ne iY 1 yh , * , i ; eran ; i we M i in iy ; i. il HM ty 7 i 4 i 5 afi | wean ie ir alae hp ne sab lle gp " ih, io t ail ewe vt Say a | Tp aera Nag The, yr a oe an! A prise ; ar : Aseanaatemiaie ne ae HNP ae at \ far He per 4 Fiplgy. bf eee | oe ie Seg ne, Cie Pury ae - : - =e. ie nuke | u ba 6 hadi | sali, «il > Seleee ae aged a8 Ale Wo Pa dag ated i | = Slerwlics «= 4asen, Dame i i - i cea a ae wan | a 1) aul grey a - oda er HAs sl 8, OS wh; 4, & ud Meee i'l eel efey beh; ful 8 | ee as ten moi ll Citta. Pay OR wel "0, MAG: 6 ui igh adh ae ah varie. aan a th As WO Validate te TALS oe. wis) = ‘he Aton pile: egy, ) @ fit 0 Pgh. Ye ey ee a ets, “Cee ae Me Bhs cg Tile | 2s Beat Wei. Yous 2 i es , on fee. 4 iret es =i ge Ma Oe AY 2 & ole fale Virco i am can a des SP HU ie Re ) AF at ties oa 9 eran all eadaadeeiahe aealaan a) 9 . sat er oe ; Ree ie ane a ae “ay t4 } a 4 7 . L : 1 . : Ce - { TTL ey Lon ; Stan A “ait ‘hana « i ; ig > sali Pm aul Met oe / vie { 4 ieag OE ee eat ; / wis : Ea Af) yo : : ; ay, ‘ia = oi ie Hi ve he, VE A, a ss 8 on 0 11) \\0 a ee oe ; ; hd AS ‘ ». A e A) , : : Re | bey ty oa ih i : t vey i: ' if) / me Til wi 4 a oa / F : 7 py ; | 1) PON aie aa | | | | ty ia mn ah Nee a LON aa SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLUME 131, NUMBER 11 (Enp oF VOLUME) THE VENTRAL INTERSEGMENTAL THORACIC: MUSCLES: OF COCKROACHES By L. E. CHADWICK Department of Entomology University of Illinois 7 wre W; THSO oO po” F ST] OnN was HINO ee (PusiicaTIon 4261) CITY OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION JANUARY 15, 1957 THE LORD BALTIMORE PRESS, INC. BALTIMORE, MD., U.S. A. THE VENTRAL INTERSEGMENTAL THORACIC MUSCLES OF COCKROACHES By L. E. CHADWICK 1 Department of Entomology University of Illinois The cockroaches, which have persisted in much their present out- ward form since the Carboniferous, are admittedly rather primitive in many structural respects. Although one dares not take for granted that their musculature also retains a primitive configuration, there is a good a priori chance that this is the case; and comparisons with other primitive types, such as Grylloblatta (Walker, 1938, 1943), the larvae of Dytiscus (Speyer, 1922) or Cybister, and larval or adult Megaloptera, to my mind leave no doubt that the muscular pattern of blattids displays many primitive characteristics. In this paper attention will be called to certain of these features, as they are seen in the ventral intersegmental musculature of the thorax. This system of muscles, though far from homogeneous morphologically, provides a convenient segment of the total thoracic musculature for analysis. It may also be regarded as itself a relatively primitive component of pterygote anatomy, for study of the more recently evolved, highly specialized flying insects shows the ventral intersegmental muscles of the thorax gripped in an evolutionary trend that has already led to the drastic reduction of these muscles and that may ultimately result in their total disappearance. In contrast, the cockroaches and other less advanced forms still exhibit a wealth of muscles in this category, and thus afford some conception of this portion of the ancestral basis from which the more adept flying insects of today have developed. Descriptions of the thoracic musculature have already been pub- lished for three blattid species: Blatta orientalis L., by Miall and Denny (1886) ; Periplaneta australasiae (L.), by Maki (1938) ; and P. americana (L.), by Carbonell (1947). Miall and Denny purposely gave only a general account ; and comparisons of the reports by Maki and Carbonell discloses more numerous and in some instances more 1 Formerly Chief, Entomology Branch, Chemical Warfare Laboratories, Army Chemical Center, Md. SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 131, NO. 11 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I striking differences among the ventral intersegmental muscles than one would ordinarily expect from members of the same order, not to say genus. Since a precise knowledge of the distribution of the ventral muscles is essential if one is to draw from them conclusions concern- ing thoracic evolution, a reinvestigation of these species was made, in the course of which it became apparent that familiarity with a wider variety of blattids would be helpful. Eventually 19 species were examined. The discussion below attempts to extract from this material information that illuminates certain important facets of the evolution of the thorax in winged insects. METHOD AND MATERIAL Specimens from culture were etherized, pinned venter-down in a wax dish, and covered with water. Under binoculars, the dorsal integument was carefully cut away from thorax and abdomen, after which the ventral system of muscles was gradually exposed by re- moving structures that interfered with the view. A few details were checked from other angles. Liberal staining with 1 percent methylene blue from time to time in the course of dissection proved helpful, and brief hardening in 70 percent alcohol was occasionally resorted to. The magnification used was 12.5 to 50 times. At various stages of dissection drawings were made to scale on squared paper with the aid of a micrometer eyepiece. P. americana (L.), Blaberus craniifer (Burm.), Blatta orientalis L., Blattella germanica (L.), and Supella supellectilium (Serv.) were available in our laboratory. Cultures of the following species were supplied through the generosity of F. H. Babers, J. H. Fales, W. L. Nutting, L. M. Roth, P. R. Ruck, C. N. Smith, and E. R. Willis: P. australasiae (L.), P. brunnea Burm., P. fuliginosa (Serv.), Blaberus giganteus (L.), Blattella vaga (Heb.), Cryptocer- cus punctulatus Scud., Diploptera dytiscoides (Serv.), Eurycotis floridana (Walk.), Leucophaea maderae (Fabr.), Nauphoeta cinerea (Oliv.), Neostylopyga rhombifolia (Stoll), Parcoblatta pennsylvanica (DeGeer), and Pycnoscelus surinamensis (L.). A single preserved specimen of Macropanesthia rhinocerus Sauss. also was dissected. The 19 species investigated were chosen mainly on the basis of avail- ability and are but a small fraction of the more than 3,500 species of cockroaches that (fide Rehn, 1951) have been described. Nymphs and adults of both sexes were examined for most species, although few differences attributable to stage or sex were found among the msucles to which the present investigation was confined. For comparison, data were obtained from the literature or from the NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 3 writer’s dissections for representatives of a number of other orders. All observations cited in this paper without a statement as to source are from my own work. Each morphologically distinct muscle was given a designation formed by hyphenating the accepted abbreviations for the skeletal parts between which the muscle is stretched. Under this system, if an attachment is segmental, the segment is identified by an arabic or roman postsubscript for the thorax or abdomen, respectively, while the designations of intersegmental structures are preceded by the appropriate arabic numeral, beginning with o for the cervical inter- segment. Exception: the customary abbreviations rcv, 2cv . . . for the cervical sclerites, and 1ax, 2ax . . . for the axillary sclerites of the wing, the latter with segmental subscripts, are retained. Cruciate muscles, with origin and insertion on opposite sides of the longitudinal body axis, are distinguished by adding X to the usual designation. The skeletal abbreviations used are for the most part those given currency by Snodgrass (1929, and numerous other publications). Examples: 2sps-epss, a muscle stretched between the second (post- mesothoracic) spina(sternite) and the metepisternum; fus-sya, the longitudinal ventral muscle from the metafurca to the second abdomi- nal sternum ; epS-cx,X, a cruciate muscle of the procoxa, with origin on the contralateral mesepisternum. A glossary of abbreviations is given at the end of the text. OBSERVATIONS AND INTERPRETATION The ventral intersegmental muscular system of the blattid thorax includes elements with primary attachments on the spinae (sps) or on the intersegmental laterosternites (ils), as well as furcal (fw) muscles that run between successive segments. This report is divided accordingly into three main sections. I. THE SPINAE Cockroaches have two authentic spinae (isps, 2sps), and in addi- tion possess in the third thoracic intersegment a common junction of serial homologs of the more anterior spinasternal muscles that lacks the median connection with the integument but obviously represents a postmetathoracic spina. This junction (“3sps”) is attached by fibrous ligaments (figs. 4, 6, 7, 9, 10, 18:27) to the bases of the metathoracic furcal arms (fus), between which it floats above the nerve cord. Comparative evidence leaves little doubt that these non- contractile ligaments, which now usually appear as fus-fus, have been 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 derived from former muscles, 3sps-fug. A true third spina, which oc- curs in some Apterygota (Maki, 1938), is known among pterygote insects only in Grylloblatta (Walker, 1938, 1943), but an arrange- ment much like that of blattids has been reported for larval Dytiscus (Speyer, 1922) and has also been seen in larval Cybister, in Zoo- termopsis (fig. 6), and in larval and adult Corydaius. Other vestiges of the third spina and its musculature have been found but not recog- nized as such by several students in a number of other insects. Alto- gether, the facts constitute strong evidence that a third spina was present in the ancestral Pterygota, and probably in early hexapods generally. More thoroughly documented, since much of the testimony is still available in a variety of living forms, is a general tendency toward loss of the remaining spinae and their associated musculature during the post-Carboniferous evolution of the pterygote thorax. Certain cockroaches, however, have gone contrary to this trend, and have experienced a prodigious extension of the first spina and, to a lesser extent, the second, in the direction of the body axis. Compare, for example, figure 1 with figure 2, or 17 with 18. This spinal elongation is related, in part, to hypertrophy of the transverse spinal musculature (Isps-epSe, 2sps-eps3) ; and it is probably no coincidence that sev- eral species that manifest this development have also exceptionally large transverse muscles of the first abdominal intersegment, syz4- sna (figs. 1,8, 11, 12: 34). My judgment that these characteristics are secondary rests partly upon the fact that they are peculiar to some blattids, being unparalleled in others and absent, so far as I know, in any other group of insects; and partly upon the presence in the mus- culature of these same cockroaches of other trends away from a primi- tive condition, such as a tendency toward loss of certain spinacoxal muscles. (See c, this section, below.) Table 1 provides a composite list of the spinasternal muscles of blattids, as these are now known, and is so arranged as to indicate the probable serial homologies. Most species possess a very large frac- tion of the total complement. The relatively few exceptions are sum- marized in footnotes to the table, and some of them are discussed briefly in the text. Included in the spinasternal musculature are (a) transverse spinal muscles; (b) spinal muscles of the preceding or succeeding furca; (c) spinal muscles of the preceding or succeed- ing coxa; (d) muscles stretched between successive spinae; and (e) spinabdominal muscles. Each of these groups is discussed under the corresponding subheading below. a. Transverse spinal muscles—The transverse muscles of cock- NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 5 TABLE 1.—Spinasternal muscles of cockroaches In this composite list, muscles that appear on the same horizontal line are considered serially homologous. Question marks indicate uncertain homologies. Symbols such as M4o, Cro3 refer to the numbers given the corresponding mus- cles in P. australasiae by Maki (1938) and in P. americana by Carbonell (1947). Each of the 19 species investigated here has all muscles shown in the table, except as stated in the notes below or in the text. For muscles without spinaster- nal attachments, see table 2. Footnote number First spina Second spina “Third spina” Bae os Isps-rils 2sps-2ils --- TDs ee Isps-epse 2sps-epss — M4o; Cro3 M74; Cr49 Deh asa Isps-fus - “3sps’’-fus MS; Cog M41 M105 (partim) Bardens se —- 2sps-fis — Cror meas del Isps-fua 2sps-fus ? “Osps’-sira aia Gros M73; Cr5z Mro4; Cig1? Rieraterayeite: s _ 2SPS-S1rA4 _ C189 Citak. st. Isps-C%1 25ps-CH2 ? fus-cxs post. rot. (partim) M24, 25; Co8 M56; C134 M88, 89; C171 Jig stevatateye ISpS-CH2 2SPS-CH3 — C105 C173 Br yates 9) Isps-2sps 2SpS-3Sps ? “3sps’-ventr. diaphr. M39; C106 M72; Cr52 M105 (partim) 1Cryptocercus has both 1,a and 1,b; other species 1,b only. The abdominal transverse muscle, $;;4-Syz4 (M112, is a serial homolog. 2 Muscle a2sps-fuz is lacking; Maki’s record (M4r) is probably an error of transcription. (See text.) 8 No comment. 4 Carbonell (1947, p. 20) describes muscle 19r in P. americana as follows: ‘‘Oblique ven- tral muscle. . . . From the base of the sternal arm to the anterior edge of the first [sic!] abdominal sternum.” 5 Abdominal insertion wholly or partly on s,;;4 in Eurycotis, Macropanesthia, Neostylopyga, and Periplaneta brunnea. (See text.) 6 Several species have two definitive muscles, Isps-cx1, one of which is probably equivalent to the eps,-fu,X of other species. (See figures and text.) In listing two furcal posterior rotators of the third coxa, Maki (1938) suggests that one of them may be serially homologous with the spinal posterior rotators of the other legs. (See text for discussion.) 7 Blaberus, Diploptera, Cryptocercus, Leucophaea, Macropanesthia, Nauphoeta, and Pyc- noscelus lack Isps-cx,; the last five genera also lack 2sps-cxry. 8 The fibers from “3sps’’ to the ventral diaphragm were not found in several species, but were possibly destroyed during dissection in some of these. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 roaches are regularly present in the first two thoracic intersegments but absent in the third, where their failure to appear is no doubt re- lated to the near obliteration of the ventral region of the first abdomi- nal segment, reduction of which is a very general and probably very early feature of pterygote reorganization. However, the transverse muscles are represented in the abdomen by the muscular attachments of the ventral diaphragm on the anterior lateral angles of the second abdominal sternum. A striking development of these abdominal fibers is seen in the series Pycnoscelus, Diploptera, Nauphoeta, Leucophaea, in which last they attain the status of a powerful trans- verse muscle, Sya-Syra (figs. 1, 8, 11, 12: 34). The function of such a muscle is unknown. Curiously, Blaberus, Cryptocercus, and Macro- panesthia (figs. 5, 10, 17), which share other peculiarities of the four genera just mentioned, do not show any tendency toward hypertrophy of the transverse muscle of the ventral diaphragm, and in this respect are more like the other species included in this study. The lateral suspensions of the ventral diaphragm are not evident in the abdomi- nal intersegments posterior to the first; and in general the ventral diaphragm of cockroaches is much less extensive than that of some other insects, e.g., phasmids and the acridid Orthoptera. Some authors have listed as transverse muscles structures such as the ligament fus-fus, whose affinities are, however, with the spina- furcal muscles. b. Spinafurcal muscles —Cockroaches all have the muscle rsps-fuy and “3sps”-fus, the latter represented, as a result of loss of the third spina, by fibrous ligaments that often appear as a single transverse band, fus-fus. A corresponding 2sps-fug does not occur in any blattid I have examined, and I believe Maki’s record of this muscle (1938, fig. 6, No. 41) in P. australasiae must rest on an error of transcrip- tion, since all cockroaches have another, larger muscle, 2sps-fu,, that is omitted from his figure and description. Miall and Denny’s (1886) statement that the muscle 2sps-fu, is inserted on the base of the first leg in B. orientalis is misleading, for the connections in B. orientalis (fig. 2: 16) are identical with those of other cockroaches ; but the description reflects Miall and Denny’s awareness of a structural difference between the prothoracic sternal arm and those of other segments, a distinction that seems to have escaped comment by most others who have investigated the muscula- ture of cockroaches. (See section 3, below.) The muscles rsps-fug and 2sps-fus are also universally present in blattids as is their possible serial homolog, “3sps’’-sya, which is here NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK fe discussed under the spinabdominal muscles. (See e, this section, below). c. Spinacoxal muscles——The first and second spinae both carry posterior rotators (or remotors) of the preceding coxa (Isps-c+1, 2sps-c%2) in all blattids examined. A corresponding muscle of the third spina is absent as such, but may be represented, as already sug- gested by Maki (1938), in the muscle fus-c1s post. rot., which fre- quently shows signs of a dual composition. In some specimens, a few of the fibers of fus-cas post. rot. appear to be continuous with those of the ligament fis-fus. Only in larval Dytiscus (Speyer, 1922), in Grylloblatta (Walker, 1938, 1943), in Zootermopsis (fig. 6: 29), and in the larvae of Cybister and Corydalus has a distinct muscle, 3sps- c#s3, been found; and in these species the muscle fus-crs post. rot., which is also present, seems to be a single band. Spinal promotors of the mesocoxa and metacoxa (Isps-c%», 2sps- cs) also occur frequently in cockroaches, as they do in other primi- tive forms. However, the mesocoxal promotor is absent in Blaberus, Diploptera, Leucophaea, Macropanesthia, Nauphoeta, and Pycnosce- lus; and the last four of these genera also lack the metathoracic homolog. Both spinal promotors are likewise missing in Cryptocercus, which shares to some extent the tendency of these genera toward hypertrophy of the transverse muscles of the first spina although it differs markedly from them in certain other respects. In some other cockroaches, e.g., in Blattella, the spinal promotors, though present, are weak. Thus, the trend toward obliteration of these muscles, which has gone far among higher orders of insects, is evident even among the Blattariae. d. Spinaspinal muscles—The muscles rsps-2sps and 2sps-“3sps” were found in all the species studied here, although they are at times weakly developed and easily overlooked; this is particularly true of 2sps-“3sps.” The first of these muscles is present also in most Or- thoptera (sensu stricto) and Mantodea, but not in other orders with the possible exception of Isoptera, where it was recorded by Fuller (1924, fig. 9, muscle n) for Termes latericius Hav. Since this muscle does not occur in other termites studied by Maki (1938) and the writer, it may be that Fuller misjudged the posterior attachment of his muscle n, which perhaps represents rsps-fu,, a muscle missing from Fuller’s account but present in all Isoptera studied by others. The muscle 2sps-“3sps” has so far been recorded only from blat- tids, where its general occurrence may be taken as one more indica- tion of primitive structure. What is probably a vestige of this muscle 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 has been found in the immature mantid, Tenodera sinensis Sauss. (fig. 7 +29). The spinaspinal muscles are the only portion of the longitudinal ventral musculature that has obviously retained its primary relation- ships in blattids. Both attachments are still unmistakeably interseg- mental. Like the other somatic muscles, the spinaspinal muscles are paired bilaterally ; but the right and left bands of rsps-2sps are often so closely appressed in the midline that they seem like a single ele- ment, and they have been so described by some authors. The pos- terior insertions of 2sps-“3sps” are usually well separated on the ligament ftts-fus (e.g., fig. 9: 23), and in some instances may even seem to be on fis at the site of attachment of the ligament. Care must be taken, therefore, not to confuse them with the usually better de- veloped spinafurcal muscles, 2sps-fus, from which they are morpho- logically distinct. ! The ligament fus-fus also serves, in several cockroaches, as a base of attachment for paired muscular strands that course posteriorly to join the meshwork of contractile fibers and membrane that consti- tutes the ventral diaphragm (figs. 2-5, 8, 11, 18: 30). These strands, “3sps’’-ventr. diaphr., may be serial homologs of the muscles /sps- 2sps, 2sps-“3sps.” I did not succeed in finding these delicate strands in all species, but could not be sure, in the cases where they seemed absent, that I had not destroyed them. In pterygote insects, there is no homolog of the spinaspinal muscles anterior to Isps; but Maki (1938) has recorded muscles that are probably homologous in the prothorax of some Apterygota. e. Spinabdominal muscles—The spinabdominal muscles of cock- roaches include only 2sps-sy4, “3sps’-srra and “3sps’-ventr. diaphr. The last two have already been mentioned in this section, b and d, above. They arise on the ligament fus-fus, and not on the base of fus as some have stated. The muscle 2sps-syr4 is characteristic of blattids, and is present in all of them I have seen, though it is weak in Leucophaea. Elsewhere, it has been recorded only from Gryl- loblatta (Walker, 1938, No. z1zb). It is interesting as an example of a muscle more than one segment in length, a type that is of infre- quent occurrence in pterygote insects. In Macropanesthia, Peri- planeta, Eurycotis, and Neostylopyga the abdominal insertion of some or all the fibers is actually on syzz4. Apparently this modification may occur readily because 2sps-sy4 is ordinarily inserted along the antecosta of syy4 dorsal to the usual longitudinal bands, syz4- Sua, With which 2sps-s;z4 is more or less continuous. Dissolution of the integumental attachment at s774 adds one segment of muscle NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 9 to the length of 2sps-syz4; and this step, to judge by various instances observed, leads to an intervening stage in which the now floating muscle is still divided by a transverse septum at the original level of attachment on sya (figs. 3, 17: 21). Subsequently, all signs of the septum are lost. Reduction of the ventral region of the first abdomi- nal segment has doubtless contributed to developments of this nature, which are not confined to cockroaches or to the particular muscle in question (cf. fig. II: 32). 2. THE INTERSEGMENTAL LATEROSTERNITES Intersegmental laterosternite (ils) is a term here introduced for sites of muscle attachment that lie at the lateral extremities of the ventral intersegmental folds. Such sites are believed to have been characteristic features of the anatomy of early arthropods in all in- tersegments. In existing forms, extensive modification of the original relationships is the rule, as will be seen below; nevertheless, recogni- tion of the presence and nature of these sites is helpful in understand- ing the manner in which the ventral musculature and associated structures have evolved. Primitively, the musculature of the ils included (1) the transverse muscles, which, with or without interruption by a median spina, stretched between the two ils of the same intersegment; (2) the out- ermost bands of the ventral longitudinal body musculature; (3) cer- tain dorsoventral muscles; and probably (4) other muscles of vari- ous types, some of which will be noted below. However, the original muscular relationships of the i/s are still not fully understood; and this fact, together with their varied fate in different lines of descent and in different parts of the body, presents the comparative morphol- ogist with many perplexing problems. Hence, it is not surprising that structures equivalent to the ils as here defined have been overlooked by some workers and variously named in different situations by others. Several have referred to them as “intersegmental pleurites,” a term which is unsatisfactory both because of the obvious sternal nature of the structures in question and because use of the name “pleurite” in reference to intersegmental elements is self-contradic- tory. Crampton (1926) avoided these objections by employing the term “furcilla.”” Unfortunately, this usage of “furcilla” seems likely to cause confusion, because the name had been applied in various other senses by earlier workers and because it suggests a nonexistent affinity with the segmental sternal apophyses, or “furcae” (fu). For these reasons, we have substituted the more accurately descriptive designation “is,” IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I1 The principal primary muscular relationships of the ils are tolera- bly well preserved in the typical abdominal intersegment, granted that a secondary extension of sclerotization has here merged the in- tersegment indistinguishably with the following segmental sternal plate, of which the former intersegment now forms the antecosta (Snodgrass, 1929). The ils are here represented in the anterolateral angles of the definitive abdominal sterna, which in many insects dis- play the muscular relationships outlined above (Ford, 1923; Maki, 1938). In the intersegments that follow each of the three thoracic seg- ments, the situation is rarely so transparent. One gains the impres- sion that, even in the most primitive forms that have come to hand, the musculature of the ils has already been subject to extensive shift- ing and reduction, while in more recent insects only a few scattered remnants suggest the original role of the ils as attachment sites for part of the longitudinal body muscles. Moreover, where the trans- verse muscles have been preserved, their lateral attachments now usually appear to be on segmental parts. In addition, we find a few muscles that originate on the thoracic ils or on their present equiva- lents, whose insertions are segmental and which have no counterparts in the legless abdomen. Equally difficult to analyze, because of the extremely varied skeletal and muscular relationships that exist in different groups, is the situa- tion in the cervical intersegment. Here one must be content for the present with the assurance that the former Oils are usually somehow represented, most often as part of one or more of the definitive cervi- cal sclerites. These problems are well illustrated in the cockroaches, in which the musculature of the is, though rich in comparison with that of more recently differentiated orders, can only be considered vestigial in relation to the inferred ancestral condition. Ventral muscles of cockroaches that appear to belong to the ils complex include (a) transverse muscles; (b) cruciate coxal and furcal muscles; (c) certain other furcal muscles; and (d) spinasternal muscles of the abdominal ils. (See tables 1 and 2.) a. Transverse muscles —The transverse muscles of the thorax ordi- narily have a median attachment on the spina, and have therefore been discussed under section I,a, above. The nature of their lateral attachments remains to be considered. As already noted, abdominal relationships support the view that the lateral attachments of the transverse muscles are morphologically intersegmental, i.e., on the ils. Comparative evidence from other arthropods and the scanty em- NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK II bryological data on insects (Heymons, 1895 ; Roonwal, 1937) justify this inference. In the postembryonic cockroach, however, the defini- tive connection in the thorax is usually with the anterior margin of the succeeding episternum, and the muscles are therefore to be desig- nated as Isps-eps,, 2sps-epss, even though Maki (1938, p. 58) de- scribes the attachment as “‘on the small sclerite before the [mes] epi- sternum” in P. australasiae. If this were all the evidence available, one would conclude that in blattids the rils and 2ils had fused with TABLE 2.—Ventral intersegmental muscles of cockroaches: muscles without spinasternal attachments Symbols such as M6, C55 refer to the numbers given the corresponding mus- cles in P. australasiae by Maki (1938) and in P. americana by Carbonell (1947). For muscles with spinasternal attachments, see table 1. Footnote number Muscle type I....Cruciate muscles Icv-C#1X €pse-Cx1X € pSe-furX Co7 C102 2....Postcoxal ligaments Tils-fus 2ils-fus 3ils-fus 3....Furcal muscles fus-tent. fus-2cv M6; C55 M7; C84 fis-fue (2bands) M38; Croo fus-fus (2bands) M71; Cr48 fu-sua (usually 3 bands) M103; C192, 193 1The 3 cruciate muscles are not serially homologous. For discussion of shifts in origin of eps,-cx,X and eps,-fu,X, see text sections 2,a and 2,b. The three muscles or their equiva- lents are present in all species examined. 2 The three postcoxal ligaments are serially homologous. For variations in their occurrence, see text section 2,c. 8 The furcal muscles are probably all serially homologous, at least in a broad sense. For variations in the abdominal insertion of fus-s;74, see text section 3. Carbonell (1947) records the abdominal attachment of muscles number 192 as on s,, in P. americana, All the furcal muscles listed are present in all species investigated. eps, and epss, respectively. The arrangement of the cruciate muscles of the profurca and procoxa in P. americana and in Cryptocercus (see below) is also consistent with this interpretation, However, the full story is not that simple, for Cryptocercus pos- esses not only muscles Isps-eps,, 2sps-epss that are clearly homol- ogous with those of other blattids, but has in addition fibrous liga- ments (figs. 10, 13: 4, 17) that run from the spinae to small sclerites in the intersegmental membrane well in advance of the episternal margin. These transverse ligaments have evidently been derived from former muscles, and their lateral attachment sites not only occupy the I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 position of true ils but also carry the dorsoventral muscles typical of these structures. Furthermore, careful dissection of most cockroaches discloses a second series of straplike ligaments, also of muscular derivation, that run from the postcoxal membranes (i.e., from the intersegments) to the furcal arms of the respective preceding segments. These liga- ments, here designated sils-fuz, etc., are inserted on the furcal arms near the seat of the furcophragmatal muscle (see figs. 2, 10: 13, 24,931). It appears, then, that in the thoracic intersegments of cockroaches the former i/s may now be represented by as many as three seemingly distinct sites: (1) the following episternum; (2) the original is; and (3) the origin of the postcoxal ligament. How this separation came about is by no means obvious. As already stated, the often straplike but still fibrous transverse ligaments rsps-rils, etc., are evidently derived from former muscles, and are even now represented in whole or in part by muscles in some species. However, again in the light of relationships found in other primitive groups (e.g., Dytiscus larva (Speyer, 1922), Corydalus larva, etc.), these transverse muscles seem to have served also (after loss of their contractile nature ?) as suspensory ligaments for a por- tion of the longitudinal ventral intersegmental musculature. Vestiges of this or an analogous arrangement are still present in the first thoracic intersegment of some cockroaches. In Cryptocercus, which in this respect is the most primitive blattid I have seen, both the ligament rsps-rils and the muscle Isps-eps2 are present and are, laterally, quite distinct (fig. 13: 4, 5). However, the mesal portion of zsps-zils, incidentally still composed of contractile tissue, is so confluent with the adjacent fibers of rsps-eps, that a separation of the two muscles in this region is hardly possible. Thus, the anterior portion of the muscle Isps-eps, could be described as “ligament-eps,.” Similarly, it is difficult to specify the origin of the cruciate profurcal muscle eps,fu,X (8), for part of its fibers arise on eps while the more ventral ones, not visible in figure 13, originate on the ligament rsps-zils, from which they run with the others to the in- sertion on the contralateral furcal apodeme fi. Variations of these relationships are exemplified in a number of other genera, viz, Periplaneta, Neostylopyga, Eurycotis, Blatta, and Blattella. In none of these is the peripheral attachment of the liga- ment rsps-rils preserved; but the central portion of the ligament is present and extends anterolaterally from the spina as a noncontractile septum on which fibers from eps, are attached and from which origi- NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 13 nate muscles (“epss”-fu,X, “eps,”-cv,X) that insert on the contra- lateral profurca and procoxa. These conditions in P. brunnea are illustrated in figure 14. Here it will be noted that the origin of the cruciate furcal muscle (8) is more central than that of the cruciate coxal muscle (9). In P. australasiae (fig. 15: 8) the more dorsal bands of the furcal muscle originate so near the midline that they appear to arise from the spina; and they are so recorded in Maki’s (1938) description. The contrasting arrangement of the corresponding muscles in P. americana is apparent in figure 16. In this species, there is no visible remnant of the ligament rsps-rils, and the cruciate muscles (8, 9) originate far laterally, on the anterior margin of eps, as they were de- scribed by Carbonell (1947). Hence the structure of P. australasiae and P. americana is superficially quite distinct. Access to intervening forms, such as P. brunnea, etc., shows, however, that in P. australasiae and P. americana we are merely confronted with rather extreme vari- ations in the arrangement of morphologically identical elements. Cryptocercus is unique among the cockroaches studied in that the most dorsal fibers of the transverse muscle of the first intersegment, Isps-epS2, continue across the body without attachment on the spina (fig. 13: 5). This development is almost certainly secondary, for the more ventral fibers of this muscle have the usual spinal con- nection. The several variants we have seen in the cruciate muscles are such that all of them could have been derived, by gradual transition, from any one chosen as a starting point. There is also at present no bar to the alternative assumption that any or all of them might suddenly have arisen de novo from each other or from an unknown basic pattern as a result of gene mutation or recombination. Therefore, a decision as to which of the existing configurations portrays the most primitive condition is not warranted on the basis of the evidence so far presented. Although Cryptocercus shows some very primitive features in the first thoracic intersegment, it is even here less primi- tive in other respects than certain other cockroaches, and should not be regarded as the prototype for the arrangement of the cruciate muscles unless independent confirmation can be produced. Other data that bear on this question are cited in this section, b, below. At the present time, it also does not seem possible to decide whether muscles such as Isps-rils and Isps-eps, are fundamentally distinct, or whether the episternal branch is no more than a hypertrophied off- shoot from an originally single transverse band. Both elements are present simultaneously in a few other insects, not all of which are I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, I3I closely related to the cockroaches. An example is shown in figure 6: 4, 5. There is some indication also, in various other primitive groups, that there existed muscles of the type rils-eps,; if so, these too may have contributed to the arrangements now seen in Dlattids, for they could conceivably account for the lateral portions of the cruciate muscles that run in some species from eps, to the transverse ligament 1sps-tils or to the septum that has replaced it. b. Cruciate coxal muscles—Mention of the cruciate coxal muscle, epSe-cx,X (9), has been made in the preceding section. This muscle is inserted together with the spinal posterior rotator Isps-cx; (7), from which it is morphologically distinct. However, species such as P. australasiae, in which the origin of “eps,”’-c7,X is far mesad, could properly be described as having two definitive spinacoxal posterior rotators, as was done by Maki (1938, fig. 6, Nos. 24, 25). In most instances these two muscles may still be distinguished by the fact that the true spinal muscle originates along the side of the spina ven- tral to the other spinal musculature, whereas the muscle equivalent to epSs-cx,X has a more dorsal origin, anterior to the transverse muscle Isps-epss. Yet the distinction is not always clear ; and in some blattids one or the other of these two muscles may even have been lost. Serial homologs of the muscle epsy-cx,;X do not occur in cock- roaches so far as is known, but homologs with origins on the ils are found in all three thoracic segments of larval Dytiscus (Speyer, 1922) and in larval Cybister. In larval Corydalus, which lacks such muscles in the first intersegment, cruciate posterior rotators of the second and third coxae originate on the corresponding furcal arms. This shift in origin is easily understood from the fact that the furcal arms are here fused with the succeeding is, evidently, as judged by conditions still found in some other Megaloptera such as Sialis spp., in consequence of sclerotization along the line of the postcoxal liga- ments 2ils-fus, 3ils-fus. One infers from the position of the cruciate coxal muscles in these primitive forms that the attachments of the cruciate muscles of cockroaches on Iils or eps, are more likely the primary ones than any of the other variants observed in blattids. If so, Cryptocercus and, for some strange reason, P. americana but not its congeners have more nearly preserved the original condition. Cruciate promotors of the first coxa have been described from a number of orders, and are apparently present in a much reduced state in all cockroaches, although on account of their delicacy they have escaped the notice of myologists. In the adult insect, which is the stage usually chosen for dissection, they are extremely slender and NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK I5 transparent. They are more easily seen, though not immediately recognizable as muscles, in the nymph, where they were first dis- covered by Scharrer (1948) as the bearers of the prothoracic glands. The glandular tissue, which encases the tenuous contractile filament and thus renders it more visible (fig. 7: 3), degenerates soon after metamorphosis, but the muscular core persists throughout life. The origin is near the anterior end of the first cervical sclerite, rcv, which for this and other reasons is to be regarded as incorporating the cervi- cal ils; and the insertion is on the proximal margin of the contralateral coxa vera, just laterad of the coxotrochantinal articulation. Cor- responding cruciate promotors of the second and third coxae have not been identified in any pterygote insect, but there is a possibility that they are represented in the usual spinal promotors Isps-cx9, 2sps-ci3. c. Lateral furcal intersegmental muscles—In cockroaches furcal muscles whose origins are on the i/s or on their present equivalents in- clude only the cruciate muscles eps -fu,X of the first intersegment ; and the three postcoxal ligaments rils-fuz, etc. The cruciate muscle has been discussed in the preceding sections. The postcoxal ligaments are often frail and transparent, and there- fore easily overlooked in dissection ; and they dissolve rapidly in alkali. These characteristics no doubt explain why the ligaments have not received more attention from morphologists, for they are quite fre- quently present in primitive insects. (See fig. 6: 13, 24, 31, and fig. 7: 24, as well as the figures of cockroaches. ) By a process that has many analogies in the evolution of the ptery- gote thorax, the postcoxal ligaments have often been replaced, in phylogeny, by apodemal growths, a course of development that cul- minates in a firm skeletal union between the furcal arm and the suc- ceeding i/s. Such unions constitute, or at least contribute to, the post- coxal bridges, whose interpretation has interested several previous students of insect morphology. Cockroaches, however, show little or no indication of the trend to- ward formation of a postcoxal bridge by sclerotization along the line of this former muscle. Only in Blaberus, of the blattids I have seen, is the distal end of the ligament 2i/s-fu, converted into a stiff, well- sclerotized apodeme ; whereas the usual course of evolution of a post- coxal bridge in other Ptergygota seems, contrariwise, to have been via sclerotization from the furcal attachment outward. In fact, the general impression left by the blattids is that their tendency is toward obliteration of these ligaments, and this tendency is increasingly manifest as one passes from the prothorax to the metathorax. All the cockroaches studied possess a fairly strong and 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I short rils-fu,; and a longer and more slender 2i/s-fug was found in all but Pycnoscelus. The presence of 3ils-fus was definitely ascer- tained only in the genera Periplaneta (4 species), Eurycotis, Blatta, and Cryptocercus. Failure to find a structure of this delicate nature is, of course, no proof of its absence ; but the observations cited never- theless do indicate quite well the tendency for these ligaments to weaken in the more posterior segments. The genera where their presence in the metathorax is most doubtful (Leucophaea, Nauphoeta, etc.) are notably those judged to have a more specialized muscular pattern on the basis of other criteria; and several of these are large insects, where such a structure, if present, should be relatively easy to find. The origins of the postcoxal ligaments of cockroaches, though clearly intersegmental, are at sites anterior and dorsal to the small sclerites indentified as the true ils by their reception in Cryptocercus of the transverse ligaments and of the usual dorsoventral muscles. In some other insects, the two sites are closer together or even indis- tinguishable, and I can offer no explanation for their separation in blattids. d. Spinasternal muscles of the abdominal ils—Only two such mus- cles have been found in cockroaches, namely 2sps-syz4 and “3sps’- Sua. The usual abdominal attachments for both are near the antero- lateral angle of the second sternum, somewhat anterior and ventral to the suspension of the ventral diaphragm. As explained above, this region of the definitive sternite is believed equivalent morphologi- cally to the thoracic ils. Identification of this attachment site with the ils renders dubious the homology, indicated as possible in table 1, of “3sps”-syr4 with the spinafurcal muscles rsps-fue, 2sps-fus; for it is very unlikely that the ils have contributed to the furcal structures of cockroaches. (See this section, c, above. ) The muscle from the second spina 2sps-syz4 clearly has no serial homolog in blattids. It is ordinarily inserted on sy74 somewhat mesad and ventrad of “3sps”-sjz4, and is thus two full segments in length. The variant attached on s;;74 has been discussed in section I,e. 3. THE FURCAE The consensus of morphologists has been that the furcae (fu) of higher insects have been produced, in phylogeny, by the approxima- tion in the ventral midline of paired segmental sternal apophyses (Weber, 1928; Snodgrass, 1929). The resulting Y-shaped structure NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK wy. consists of the infolded furcabasis and the laterally extended furcal arms. In cockroaches, as in other primitive forms, right and left apophyses remain separate. For this reason, purists avoid applying the term “furca” to them, but for convenience we shall continue to refer to them as the furcae or furcal arms, with which they are homologous. Despite the seemingly incontestable segmental nature of these apophyses, they nevertheless carry a large fraction of the surviving longitudinal ventral intersegmental musculature in all ptery- gote insects. This situation poses a contradiction, long recognized and accepted by students of thoracic structure, that has never been satis- factorily resolved (cf. Snodgrass, 1929). Weber (1928) surmised that the present furcal intersegmental mus- cles had been derived from spinasternal muscles. He proposed that, as the furcal arms were gradually elevated in phylogeny, they inter- cepted the spinasternal muscles, which thereupon acquired furcal at- tachments and lost their primary connections with the spinae. This hypothesis, which regards the furcal muscles as replacements for the spinasternal muscles, is clearly untenable in the face of the presence of the usual complement of furcal intersegmental muscles in all those primitive forms, such as larval Dytiscus (Speyer, 1922) or Cybister, larval Corydalus, and the cockroaches, which still retain an extensive spinasternal musculature, including (in the blattids) both spinaspinal muscles Isps-2sps, 2sps-“3sps.’ Conceivably, the rising furcal arms could have intercepted some of the more lateral bands of the primary ventral longitudinal intersegmental muscles, for instance those at- tached on or below the transverse ligaments, Isps-rils, etc. ; but even this modification of Weber’s hypothesis is unconvincing in the absence of any known situation in insects where interception of a muscle by a skeletal element has led to the development of an attachment be- tween the two. Moreover, the data of Carpentier, Barlet, and others (see Barlet, 1954, for references) show that the essential features of the furcal complex exist in the Apterygota, which also possess an extensive array of muscles homologous with the spinasternal muscles of higher forms. Any notion that the furcal longitudinal muscles have arisen in the Pterygota through transfer of muscles from some other category must therefore be abandoned. How then can they be ac- counted for? An answer may be approached, we believe, through realization that the principal endoskeletal structures of insects and other arthropods have all developed as the result of sclerotization along the course of former muscles, and that the present sternal arms are of this nature. Although the genesis of certain endoskeletal structures lies so far in 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 the past that it will probably be impossible forever to document the details of the process in these instances, there are many other cases where the course of evolution can be deduced with reasonable cer- tainty from comparisons of existing forms. The complex endosternal structures of the Apterygota furnish a number of examples, for, as described by the Belgian authors cited above, many endosternal ele- ments that are ligamentous in one species or group are still repre- sented by functional muscles or by apparently degenerate muscles in others. In the opinion of the present writer, yet other parts of the endosternum that are invariably ligamentous in the apterygote species so far studied are homologous with muscles, such as the transverse muscles, that persist as contractile elements in some primitive Ptery- gota and as ligaments in others. Another clearcut set of examples of the replacement of muscles by endoskeletal structures is found in the later history of the sternal arms themselves, for instance in the de- velopment of the furcopleural fusion, which has occurred inde- pendently in numerous lines of descent. Here the process can be fol- lowed in some detail through several series of intermediate stages provided by existing forms. As a generalization we offer the hypothesis that all such endoskele- tal developments owe their inception to other structural or functional changes that have limited freedom of movement at the insertions of certain muscles. These muscles, deprived of their original effective- ness as contractile organs, are doomed to disappear unless they happen to retain some value in the role of static supports or braces. Further- more, the organism evidently finds it more economical to construct the braces it requires from other than contractile tissue, which cannot resist compression and which can maintain tension only through a continuous expenditure of energy, so that replacement of bracing mus- cles or tensors by noncontractile ligaments or by stiffer sclerotized apodemes is the usual evolutionary pattern. In our view, structures originating in this manner constitute the primary endoskeletal rudi- ments. Once established, these may be variously molded in later evo- lution in accordance with the mechanical requirements they are called upon to fill; and in the course of such modification their original derivation from muscles may be almost wholly obscured. Returning to the narrower problem of the nature of the sternal arms and their longitudinal musculature, we may point out that the arms are represented in the Apterygota by ligamentous straps that con- nect the thoracic endosterna, which are mainly intersegmental in char- acter, with the respective preceding segmental sternal regions (refer- ences in Barlet, 1954). In these insects the endosterna provide the NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 19 attachment sites for almost all the ventral musculature, including, of course, the usual longitudinal intersegmental muscles. We have al- ready indicated our belief that the entire endosternal complex, which is ligamentous in consistency, is of muscular derivation ; and we sug- gest here that the sternal ligaments are merely another example of transformed muscles. We may then regard the endosternum schemati- cally as a point of junction of various intersegmental muscles, among which are the usual longitudinal bands and a muscle to the preceding segmental sternum. The configuration thus summarized is, however, exactly what is seen in the musculature associated with the sternal arms of pterygote insects. True, the number of elements that impinge upon this focus is less than in the Apterygota; but those elements that do occur in the Pterygota all have their counterparts in muscles that are attached on the apterygote endosternum in proximity to the attachments of the sternal ligaments, or in similarly directed portions of the endo- sternum itself. Only the fact that the sternal arm of Pterygota is usually a heavily sclerotized ingrowth of the ventral body wall gives the superficial impression of a fundamental difference between the two subclasses. In cockroaches, even this distinction breaks down; for in the prothorax of blattids the paired furcal pits do not give rise at once to sclerotized apodemes, as they do in the mesothorax and metathorax, or in the prothorax of most Pterygota. Instead, there extends inward from the pit a flexible, fibrous ligament that connects with the apex of a sclerotic bar whose other end articulates firmly with the pleural apodeme. Upon this bar, at or near its junction with the sternal liga- ment, are inserted the usual muscles of the furcal complex. On the basis of these facts and the considerations outlined above, we suggest that the sternal arms of pterygote insects represent muscles that formerly ran from the segmental sternal region posteriorly to a common junction of various other intersegmental muscles, including the forerunners of the present longitudinal furcal muscles. In the course of evolution, the sternal muscles were replaced first by non- contractile ligaments, a condition still manifest in the Apterygota and in the prothorax of blattids, and finally by sclerotized apodemes, the form in which they now appear in the pterothorax of cockroaches and in all three thoracic segments of most winged insects. These changes in the sternojunctional muscle have not altered the morphological rela- tionships of the other muscles attached at the junction, which may still be regarded as an intersegmental locus in the morphological sense. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I There is thus no problem of explaining a shift of their attachments to a segmental site, for no shift has occurred. Apart from the structure of the profurca, with its connotations for the evolution of the furcal structures of pterygote insects in general, there is little that is remarkable about the sternal apophyses and their musculature in cockroaches that has not already been touched on in preceding sections. The ventral furcal intersegmental muscles found in the Blattariae may be classified as (a) spinafurcal muscles; (b) furcal muscles from the ils; and (c) furcofurcal muscles, includ- ing muscles with furcal origins and insertions in the head, neck, or abdomen. If the suggestion offered above is correct, that the definitive furcal apophysis is partly of intersegmental nature, a number of other muscles with furcal attachments, such as those of the appendage, may also be primarily intersegmental. However, further work is needed on the details of such relationships, and it seems best to leave them for future consideration. Except for muscles with spinasternal attachments (table 1), the ventral furcal intersegmental muscles of cockroaches are listed in table 2. Furcal muscles from the spinae and i/s have been discussed above, particularly under sections 1,b, 2,a, and 2,c. Like these muscles, the furcofurcal muscles of blattids can be homologized in considerable detail with those of comparable location in other insects. Readily dis- tinguished in most cockroaches are a usually slender mesal band and a more massive lateral band of both fw,-fu2 and fus-fus. Components probably homologous with each of these bands can also be identified in many blattids in the muscle fi3-sr74, which often includes an addi- tional more ventral group of fibers. These subdivisions, ordinarily lumped together in descriptions, seem to possess a fair degree of constancy in a number of insect orders, and may be of significance in future more detailed comparative studies. The fact that the furcabdominal muscles are inserted on the second (never on the first) abdominal sternum, is what would be expected if the furcal attachment is really intersegmental, as has been argued above. Morphologically, these muscles still run from the third thoracic intersegment to the first abdominal intersegment, and have neither lost nor shifted their original attachment sites. However, just as the muscle 2sps-syz4 has become 2sps-syzz4 in some species (sec- tion 1,e), so is a portion of fts-sy4 found, at times, as fus- Sra, aS a result of an analogous development (figs. 1, 5, 8, II, 12, w7ose) NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 21 The profurca bears two anteriorly directed longitudinal ventral intersegmental muscles. The stronger, consisting of two or more bands and serially homologous with the muscles fu;-ftte, fue-fus, passes into the head to insert on the tentorium. This muscle, fu,-tent (figs. 1-18: r), is commonly considered to be more than one segment in length (Snodgrass, 1935, p. 159). The weaker, usually a thin, flat strap of somewhat degenerate appearance, has a more ventral origin on fu; (figs. 4, 5, 9: 2) and is inserted on the mesal lobe of the ipselateral second cervical sclerite. Possibly this site should be re- ferred to the Oils, which are certainly included in the first cervical sclerites, rcv, of which the 2cv may be merely subdivisions (Cramp- ton, 1926) ; but this question cannot be settled until the constitution and muscular connections of the various cervical structures of insects are better understood. GENERAL DISCUSSION Reinvestigation of the ventral intersegmental muscles of the three cockroaches previously studied by others has shown each of the earlier accounts deficient in some respects. The defects are mostly errors of omission. Thus, none of the earlier investigators noted the muscle Icv-c+,X or the three postcoxal ligaments Z1/s-fu;, etc. That they did not is understandable, for these are hardly muscles in a func- tional sense, even though the cervicocoxal “muscle” does have a con- tractile core a few microns in diameter. Carbonell’s (1947) exclusion of the ligament fis-fus3(=3sps-fus) and of the bands “3sps’’-ventr. diaphr. from his description is likely to have been on similar grounds ; for in most other details his depiction of P. americana is accurate and complete. However, unless Maki’s (1938) specimens of P. australasiae differed radically from those of this species available to the writer, one must reject Maki’s assertions that the muscles Isps-fuy, Isps-cx%2, 2sps-fuy, 2sps-cx3, and 2sps-sy4 are absent, and that a muscle, 2ps-fu2, unknown in any other cockroach, is present. Miall and Denny (1886) stated explicitly that they had not given a complete account of the muscles of B. orientalis, so that there is no cause for surprise in the fact that this species has numerous muscles unmentioned in their description. When these few corrections have been made, it is seen that all species of cockroaches thus far studied have nearly identical com- plements of ventral intersegmental muscles. The spinasternal pro- motors of the coxae, Isps-cx, and 2sps-cx 3, are absent from some species and show signs of weakness in others. The metathoracic post- 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I coxal ligament 3is-fus is either lacking or very frail in several genera. Cryptocercus alone of the species investigated possesses muscles as yet found in no other blattid ; these are the semiligamentous transverse bands rsps-rils and 2sps-2ils, parts of which seem to have survived as septa in some other cockroaches. With these minor exceptions, the differences among the several species are merely variations in rela- tive size and proportions of the various muscles, or occasionally in their position. Such differences, some of them quite conspicuous, ob- viously indicate shifts in functional emphasis, although in most in- stances the details of their interpretation from this viewpoint are obscure. In contrast with most modern insects, the cockroaches enjoy a relatively rich ventral intersegmental musculature. Some authorities would be inclined, perhaps, to regard this as a consequence of secondary reduction of sclerotization in the ventral regions of the blattid thorax ; but acceptance of this view would make it very hard to account for the presence of homologous muscles in a number of other groups in which the thorax is extensively sclerotized. There is no muscle recorded in this paper for which either a direct or a serial homolog has not been found in at least one other order of pterygote insects, and most of them are known from several. Coupled with the fact that those other orders that display a similar degree of com- plexity in the ventral intersegmental musculature are the ones con- sidered highly primitive in various other respects, the evidence seems more consistently interpreted in the conclusion that the cockroaches also are primitive in this feature, and that the primitive state of the ventral intersegmental muscles was a complex one. As already indi- cated in the introduction, the various structural patterns preserved for our scrutiny among the more recently evolved orders of insects, which constitute a progressive series of specializations toward greater efficiency in flight, show that improvement in the flight mechanism has been accompanied regularly by reduction in the ventral inter- segmental thoracic musculature. These facts also favor the view that the possession of numerous discrete muscles in this category is a primitive characteristic. The results of a more comprehensive inquiry into these problems will be reported elsewhere. SUMMARY 1. A comparative study was made of the ventral intersegmental musculature of the thorax in 19 species of cockroaches. The observa- tions produced a few corrections, mainly additions, to earlier de- NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 23 scriptions of 3 of these species. In general, there are very minor differences among the species in respect to the presence or absence of individual muscles in the category studied, although there are numerous differences, some conspicuous, in the relative size and pro- portions of the various muscles. 2. The present ventral intersegmental thoracic muscles had their primary attachments on the spinae, on the intersegmental latero- sternites, or on the forerunners of the furcal apophyses. Cockroaches still have two typical spinae and definite vestiges of a third. They possess an extensive spinasternal musculature. Remnants of the musculature of the intersegmental laterosternites are present, but some of these muscles now have segmental attachments, and others are represented by noncontractile ligaments. The longitudinal furcal muscles are equivalent to those of other pterygote insects. 3. Attention is called to the postcoxal ligaments that run between the furcal apophysis and the immediately following intersegmental laterosternite in each of the thoracic intersegments, and to the sig- nificance of these former muscles in the development of the postcoxal bridges of higher insects. 4. Evidence and arguments are presented for the hypothesis that the furcal apophyses represent former muscles that have been replaced in phylogeny by sclerotized apodemes. It is suggested that one at- tachment of these muscles was on the segmental sternum, the other at a common intersegmental junction of several other muscles, among them elements of the longitudinal ventral group. Loss of movement at the sternal insertion led first to transformation of the sterno- junctional muscle into a fibrous ligament and eventually to the scle- rotization of the ligament. Analogous events have occurred fre- quently in the evolution of the pterygote thorax. The blattid prothorax exemplifies a stage in the evolution of the typical furcal apophysis when the postulated sternal muscle was still in a liga- mentous condition. The subsequent sclerotization of this ligament, which has occurred in the other thoracic segments of cockroaches and in all three segments of most Pterygota, in no way alters the morpho- logical relationships of the other muscles inserted upon its central end; morphologically, then, the present attachments of the longitudi- nal ventral muscles on the furcal arms are still intersegmental, and it is therefore unnecessary to invent mechanisms whereby they might have been shifted from an intersegmental to a segmental site of attachment. 5. The ventral intersegmental thoracic musculature of cockroaches is rich in number of elements, compared to that of more recently 24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I31 evolved groups. However, direct or serial homologs of all the ele- ments occur in one or more other orders of winged insects. The most extensive complements of these muscles are found in those forms, such as larval dytiscids, Grylloblattodea, and Megaloptera, that are regarded as primitive on the basis of other criteria. It is concluded, therefore, that the blattids also are primitive in respect to the ventral interseg- mental muscles ; and that possession of a rich ventral intersegmental musculature was characteristic of the early Pterygota. ACKNOWLEDGMENT The writer is grateful to Dr. R. E. Snodgrass for helpful discussion of several questions considered in this paper. GLOSSARY OF ABBREVIATIONS CR aaah eines ehisaesie cervical sclerite Cae NG Soehion tue acis ls coxa OPS mal ccatate eihate Oehatere! tev .. episternum | (OIG A TiS CRS AIG ae oo furca, furcal arm, segmental sternal apophysis AUS aE Misia io tebe chet carats intersegmental laterosternite Dy vericic eisgesiite atelier phragma, or the primary dorsal intersegmental fold from which the phragma is derived DOSET Ob tole cisions Sees posterior rotator, a functional designation used to distinguish certain leg muscles SHER Saas sath eee segmental sternum SEP Een alate cistotoinioisietais tothe septum OS te aca chan pce mitt spinasternite or spina Pe ie ofetcistaciecraiete wiorslueteleie.s segmental tergum COME Ao aeicte eee a hee tentorium Vente APMP) Fs cece aes ventral diaphragm EN ee eer iat cule poor shat cruciate, used of a muscle whose origin and insertion are on opposite sides of the midline For the way in which these abbreviations are compounded into designations of muscles, see section on Method and Material in text. REFERENCES BaRteT, J. 1954. Morphologie du thorax de Lepisma saccharina L. (Aptérygote Thy- sanoure). II.—Musculature (2™° partie). Bull. Ann. Soc. Ent. Belgique, vol. 90, pp. 299-321. CARBONELL, C. S. 1947. The thoracic muscles of the cockroach Periplaneta americana (L.). Smithsonian Misc. Coll., vol. 107, No. 2, 23 pp., 8 pls. Crampton, G. C. 1926. A comparison of the neck and prothoracic sclerites throughout the orders of insects from the standpoint of phylogeny. Trans. Amer. Ent. Soc., vol. 52, pp. 199-248. NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 25 Forp, N. 1923. A comparative study of the abdominal musculature of orthopteroid insects. Trans. Roy. Can. Inst., vol. 14, pp. 207-319, 17 pls. Futter, C. 1924. The thorax and abdomen of winged termites. Union of South Africa Dept. Agr. Ent. Mem., vol. 2, pp. 49-78. Heymons, R. 1895. Die Embryonalentwickelung von Dermapteren und Orthopteren unter besonderer Beriicksichtigung der Keimblatterbildung. viii + 134 pp., 12 pls. Jena. MaAkgI, T. 1938. Studies on the thoracic musculature of insects. Mem. Fac. Sci. Agr. Taihoku Imp. Univ., vol. 24, No. 1, 343 pp., 17 pls. Mratt, L. C., and Denny, A. 1886. The structure and life history of the cockroach (Periplaneta ori- entalis). 224 pp. London. REHN, J. W. H. 1951. Classification of the Blattaria as indicated by their wings (Orthop- tera). Mem. Amer. Ent. Soc., No. 14, 134 pp., 13 pls. Roonwat, M. L. 1937. Studies on the embryology of the African migratory locust, Locusta migratoria migratorioides Reiche and Frm. II.—Organogeny. Philos. Trans. Roy. Soc. London, B, vol. 227, pp. 175-244. ScHARRER, B. 1948. The prothoracic glands of Leucophaea maderae (Orthoptera). Biol. Bull., vol. 95, pp. 186-108. Snoperass, R. E. 1929. The thoracic mechanism of a grasshopper, and its antecedents. Smith- sonian Misc. Coll., vol. 82, No. 2, 111 pp. 1935. Principles of insect morphology. ix + 667 pp. New York. Speyer, W. 1922. Die Muskulatur der Larve von Dytiscus marginalis. Ein Beitrag zur Morphologie des Insektenkérpers. Zeitschr. wiss. Zool., vol. 119, PP. 423-492. WALKER, E. M. 1938. On the anatomy of Grylloblatta campodeiformis Walker. 3. Exoskele- ton and musculature of the neck and thorax. Ann. Ent. Soc. Amer., vol. 31, pp. 538-640. 1943. On the anatomy of Grylloblatta campodeiformis Walker. 4. Exo- skeleton and musculature of the abdomen. Ann. Ent. Soc. Amer., vol. 36, pp. 681-706. Weter, H. 1928. Die Gliederung der Sternopleuralregion des Lepidopterenthorax. Zeitschr. wiss. Zool., vol. 131, pp. 181-254. 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I EXPLANATION OF FIGURES All muscles figured are numbered uniformly according to the list below. Instances in which the definitive attachments differ from those given in the list have been discussed fully in the text. The arrangement of ventral muscles in certain species is such that not all of them can be shown in a single drawing. However, all cockroaches studied have all muscles given in the list, whether the muscles appear in the figures or not, except as noted under tables 1 and 2 or in the text. The termite Zootermopis (fig. 6) and the mantid Tenodera (fig. 7) have only the ventral intersegmental muscles shown in the drawings. In a few instances different levels of dissection have been shown in different parts of the same figure; this does not imply an absence of the usual bilateral symmetry. To assist in orientation, some figures contain a few muscles that do not belong to the ventral intersegmental category. The scale indication represents I mm. NUMBERING OF MUSCLES IN FIGURES I-18 Number Muscle Number Muscle Number Muscle 1 fus-tent. 13 rils-fitx 25 fus-2ph 2 fur-2cv 14 fiti-Iph 26 ftts-fuls 3 Icv-cxiX 15 fits-fus 27 “3sps’-fus 4 Isps-trils 16 2sps-fus 28 “3sps”-sira 5 Isps-epss 17 2sps-2ils 29 fils-c2s post. rot.* 6 Isps-fus 18 2sps-epss 30 “3sps’-ventr. diaphr. 7 Isps-Cx1 IQ 2Sps-CH2 31 3tls-fus 8 epse-fusX * 20 2sps-fus 32 fus-Srra * Q epse-cx1X * 21 2sps-sua * 33. fus-3ph (or -tra) 10 Isps-fits 22 25SpS-CHs 34 Sura-Sura Il ISpPS-CHs 23 2SpS-3ZSps 35 Sua-Sua I2 Isps-2sps 24 «2ils-fus * Variants from these attachments are discussed in the text. NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK WW ZK A hb SS \ 30 a’ NWS Wi ZA sept =! JONY4\ 34 ‘Sina ————— 4. Rage i ra wy -7+9 15 — Aut o- NN rot 16- ea ‘thts yy 3 NS 3l 2? Ss ‘NY 734 Fics. 1-4.—1, Leucophaea maderae (Fabr.), male. 2, Blatta orientalis L., male nymph. 3, Neostylopyga rhombifolia (Stoll), male. 4, Blattella vaga (Heb.), female. : th, i) it Sh - SSS eS -- “img So tee an Mh AN (I aI AMP AA 22 20- PA ah — — 23 Ava Xf ( \ DN (OS 2t di =i) ? ir -I9 Fe nent MWA Ee fle NT NS AY hha He Wi Aeta SEE AMEE pe Wa AAAS al W EO) AL- NY i \ fy SO) AGS BIN SAAa eT] R S A SS 57, Fics. 5-8—Blaberus craniifer (Burm.), nymph. 6, Zootermopsis angusticollis (Hagen), worker (Isoptera). 7, Tenodera sinensis Sauss.. nymph (Mantodea). The prothorax is shown on the left, the mesothorax and metathorax on the right. The portion of the prothorax posterior to the sternal arms, which does not carry any ventral intersegmental muscles, has been omitted. Glandular tissue that invests the muscles Icv-cv:X (3) is shown in solid black. 8, Nauphoeta cinerea (Oliv.), nymph. NO. II THORACIC MUSCLES OF COCKROACHES—CHADWICK 29 {NA Lin ~. ae ISN 41). 6+8 Se - 5 ~ha i ee iN -!0 SK \ ! GPS _ 24 Ae 9 2 a 20- 4 Hf he es o ZN ip me oT LZ “26 Sia A i wre Sw Wa zw sie is Aye 15 hy Wyaz NY == ave) Fics. 9-12.—9, Supella supellectilium (Serv.), female. 10, Cryptocercus punc- tulatus Scud., female. 11, Pycnoscelus surinamensis (L.), female. 12, Diploptera dytiscoides (Serv.), nymph. 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. T3t TIS i Lg i —— =a. i2--Woonoa DT Toe -16 d- 7+ 9- (Zig 13 I5~ Wi! 7 YP ca wat} WW ktie rie 6-_ Shika 3- AY. i i ae SAINI PZ 8 I5S+16-- 7 wii ll ~ 9 6-- =2 eee Ca 2 B& ie ae WS-- 215 2 i. Y A oa VE ZZ -I9 26 QY- - 20 ae Nl = ree Bb 21----H NG ----20 ei Aix 8 a ik dy NS ----2l