NEW MAMMALS. 89 



preservation of the smaller herbivorous mammals, that it may still be possible to secure recent 

 specimens both of this new rodent and also of the new species next described. 



Agouti thomasi, sp. nov. 

 Plate XXXVIII, figures 6-8; Plate XXXIX, figures 1-3. 



Bones of food animals obtained from burial caves and middens include skeletal parts 

 of eight rodents of the genus Agouti. Some of the material is very fragmentary, as might 

 be expected considering the manner in which it was preserved; therefore, although it is 

 quite possible that all these animals represent the same species, I cannot prove this to 

 be the fact. It is evident, however, that the largest and best specimen of the series, a 

 fairly well-preserved adult (aged) skull found in a small kitchen-midden near the eastern 

 limit of the city, is distinct from all species hitherto described under the genus Agouti. I 

 have therefore selected this skull (Ost Coll. 3327) as the type of a new species, which 

 I have the honor to name Agouti thomasi after Mr. Oldfield Thomas of the British Museum 

 (Natural History). 



I have for comparison no material of A. tacsanowskii, M. J. Stolzmann, but the excellent 

 description and illustrations of that author and the comprehensive table of measurements 

 compiled by him show at once that the new type skull does not belong to the species he 

 described from Equador (Proc. Zool. Soc. London, 1885, page. 161). The new skull chffers 

 from A. taczanowskii principally in the following details: 



It is longer and has a much smaller zygomatic breadth, while the greatest height of 

 the zygoma is also very much less. The flare or splay of the posterior wings of the zygomata 

 in the new skull is more pronounced than in A. taczanowskii, and the maximum zygomatic 

 breadth is conspicuously greater than the zygomatic breadth taken opposite the second 

 molar, while in A. tacsanozuskii the difference between these two diameters is insignificant, 

 the latter species approaching A. paca in this respect. I may add that a skull of A. paca 

 (Ost. Coll. 120) is actually wider opposite the second molar than at the posterior angles 

 of the zygomata — a peculiarity of form that is correlated with the immense size of the 

 lateral oral capsules. 



The surface is quite smooth, even the lower borders of the zygomata being entirely free 

 from the corrugations so characteristic of A. paca and A. taczanowskii. This total lack of 

 superficial corrugation on the new type skull, which has somewhat thin inferior zygomatic 

 margins, leads me to suppose that it is female, since a younger and slightly smaller skull 

 (Ost. Coll. 3316), Plate XXXVIII, figures 6, 7, from Cave 13, and also a fragment Of 

 a still larger skull from Cave 9 exhibit moderately roughened surfaces near the lower 

 borders of the zygomata, the inferior edges of these arches being stouter than in the type 

 skull. I can see no reason why this last difference should not have here the same sexual 

 significance that it has in the case of human crania. As no lower jaw was found with the 

 type skull, an imperfect mandible from Cave 99, presumably of the same species, is shown 

 in Plate XXXVIII, figure 8. 



Mr. Oldfield Thomas' Agouti sicrrcc from Venezuela is the species that the new skull 

 resembles most closely, as will appear from the accompanying table, in which are arranged 

 in parallel columns, the measurements of A. sierrcc (Annals and Magazine of Natural 

 History, 1905, 7th Sen, Vol. XV, page 589) and those of the new Peruvian species. 



