84 BULLETIN 16 9, UNITED STATES NATIONAL MUSEUM 



Genus PTILODUS Cope, 1881 



PTILODUS MONTANUS Douglass 



Figure 8 



Ptilodus montanus Douglass, 1908, p. 14; Gidley, 1909, p. 615; Granger and 



Simpson, 1929, p. 632. 

 Ptilodus gracilis Gidley, 1909, p. 616; Granger and Simpson, 1929, p. 633; non 



Ptilodus gracilis (Marsh, 1889) Osborn, 1893. 

 Ptilodus admirabilis Hay, 1930, p. 380, to replace Ptilodus gracilis Gidley, 1909, 



non Osborn, 1893. 



Type. — Carnegie Museum no. 1673, left lower jaw with P4 and Mi. 

 Collected by A. C. Silberling. 



TyjJe 0/ Ptilodus gracilis Gidley. — U.S.N.M. no. 6076, skull, jaws, 

 and partial skeleton. Collected by A. C. Silberling, 



Horizon and locality. — Fort Union no. 2, Middle Paleocene horizons, 

 Crazy Mountain Field, Mont.. 



Diagnosis. — Length P4, mean 8.0 ±0.07, standard deviation 0.42 ± 

 0.05. Length Mj, mean 3.4 ±0.07, standard deviation 0.20 ±0.05. 

 Length P*, mean 5.5±0.11, standard deviation 0.38±0.08. Width 

 P^ mean 2.6 ±0.06, standard deviation 0.22 ±0.04. Ratio length 

 P4 : length Mi, mean 2.3 ±0.02, standard deviation 0.04±0.0L Ratio 

 length Ml : width Mj, mean 2.0±0.04, standard deviation 0.13 ± 

 0.03. Serrations P4 13-15, mode 14. Cusps P^ 4-7, mode 6. Outer 

 cusps P* 0-3, mode 0. Inner cusps P^ 9-10, mode 9. Outer cusps 

 M^ 7-9, mode 8. Cusps Mi external 5-6, mode 6, internal 4-5, 

 mode 4. 



Discussion. — This is far the commonest single species in the fauna. 

 The type happens to be almost exactly at the mean or mode for every 

 character that it shows, and hence it is extraordinarily well fitted to 

 be the type, although this was, of course, accidental, as it was almost 

 unique when described. Gidley distinguished his Ptilodus gracilis as 

 being slightly smaller than P. montoMus, lower jaw far more slender, 

 and five outer cusps on Mi as against six in P. montanus. The other 

 characters given were, as Gidley recognized, not comparable with or 

 not distinctive from P. montanus. Mi appears to me to have six 

 external cusps, rather obscured by wear.^^ The slight size distinctions 

 are not valid specific characters, for now that the whole collection 

 can be compared it is seen that P. gracilis Gidley falls definitely 

 within the range of P. montanus in every respect. It happens to 

 be one of the smallest specimens of tliis species, and this unfortunate 

 chance, not recognizable as such when he wrote his preliminary 

 paper, misled Gidley into thinking it representative of a separate 

 species. 



3' The presence of only five would not necessarily be distinctive anyway, as one specimen, surely of 

 montanus, has only five and two others have five large cusps and one small and indistinct. 



