2 PROCEEDENGS OF THE NATIONAL MUSEUM vol. los 



cott, which he regarded as the "calcified chordal sheath" of a fish 

 alhed to the bradyodonts, but which, as later researches have clearly 

 shown (Dean, 1906, pp. 132-135; Bryant, 1936, p. 410; Flower, 1952, 

 pp. 516-517), cannot possibly be of vertebrate origin. The Harding 

 Sandstone vertebrate remains have been dealt with by Jaekel (in an 

 appendix to Walcott's paper of 1892) and subsequently by, among 

 others, Vaillant (1902), Eastman (1907, p. 33; 1917, pp. 236-239, pi. 

 12, figs. 5, 6), Woodu^ard (1921, p. 179), Stensio (1927, pp. 314-315, 

 333), Stetson (1931, p. 153), Bryant (1936), Berg (1940, pp. 107-108, 

 360-361), Kvam (1946, pp. 19-20, fig. 1), Gross (1950, p. 73; 1954, p. 

 80, pi. 3, fig. 2, pi. 5, figs. 3, 4, 7), 0rvig (1951, pp. 381-382, 387, 393, 

 415, 433, fig. 22 b, pi. 3, figs. 3, 4), Gregory (1951, vol. 1, pp. 102- 

 104, vol. 2, figs. 6.2 (a), 6.3, 6.4), Denison (1956), J. D. Robertson 

 (1957), and James (1957, p. 9, pi. 1). A description of Astrasjns 

 desiderata, based on the only fairly complete carapace of this form 

 known so far,^ mil be given in a forthcoming paper by the present 

 writer (0rvig, in MS., b). Various notes on the invertebrate fauna 

 and/or the stratigraphy of the Harding Sandstone have been given by 

 Walcott (1892), Darton (1906b), S. R. Kirk (1929), Miller (1930), 

 Stauffer (1930, p. 83), E. Kirk (1930), Branson and Mehl (1933); 

 Behre and Johnson (1933), Ulrich (1938), Johnson (1944, pp. 320-322), 

 Flower (1952), Frederickson and Pollack (1952), Twenhofel et al. 

 (1954), Sweet (1954; 1955), and Denison (1956, pp. 368-369). This 

 formation is now generally believed to be of Middle Ordovician 

 (Trenton) age. 



Vertebrates to some extent resembling those of the Harding Sand- 

 stone have long been known to occur in certain sandstone beds 

 beneath the Bighorn Dolomite of north-central W3^oming and in the 

 lower sandstone and silts tone members of the White wood formation of 

 the Black Hills region in South Dakota, but none of this material has 

 yet been properly described (see Darton, 1906a, p. 29; 1906b, pp. 

 550-551, fig. 3; 1909; Cockerell, 1913, p. 247; E. Kirk, 1930; Miller, 

 1930, p. 206; Romer and Grove, 1935, pp. 810-811 ; Furnish, Barrag}^, 

 and Miller, 1936, pp. 1335-1338, pi. 2, figs. 14-16; Amsden and Miller, 

 1942, p. 304; Miller, Cullison, and Youngquist, 1947, p. 31; McCoy, 

 1952; Denison, 1956, p. 367). Dermal elements of Thelodontida and 

 Osteostraci (Ateleasjns?) have been recorded from strata of a presumed 

 lower Middle Ordovician age in the Bighorn Mountains of Wyoming 

 (Tieje, 1924), but everything seems to indicate that those fossils are 

 incorrectly determined. A new occurrence of Ordovician vertebrates 

 has recently been detected farther north, in the sandstone and shale 

 of the Winnipeg formation of the Williston Basin. This latter fauna, 



'This specimen (USNM 8121), mentioned already by Walcott (1892, footnote on p. 167) and later de- 

 scribed by Eastman (1917, pp. 2.38-239, pi. 12, Ars. 5, 6) and by Bryant (1936, pp. 416-417, pi. 1), is of par- 

 ticular Interest in that It shows part of the lateral line canal system on the anterior part of the carapace. 



