We have already remarked that the last two chapters of the first book of the _Physionomia_ suggest that in compiling them the author had before him an Arabic treatise on Natural History. A natural conjecture leads us further to suppose that this may have been the original from which he translated the _De Animalibus ad Caesarem_, and this idea becomes a certainty when we pursue the comparison a little more closely. Take for example this curious passage from the _Physionomia_ (Book I. chap, ii.): ‘Incipiunt pili paulatim oriri in pectine unitas quorum dicitur femur … item sibi vox mutatur.’ Its obscurity disappears when we confront it with the corresponding words in the _De Animalibus ad Caesarem_, and thus discover what was no doubt the original source from which Scot derived it: ‘Incipiunt pili oriri in pectore _Kameon alkaratoki_, et in isto tempore mutatur vox eius.’[85] There is no need to extend the comparison any further than this significant passage. Doubt may arise regarding the depth and accuracy of Scot’s knowledge of the Arabic tongue, the nature of the text that lay before him, or the reason he may have had for retaining foreign words in the one version which he translated in the other; but surely this may be regarded as now clearly established, that some part of the first book of the _Physionomia_ was derived by compilation from the same text which appeared in a Latin dress as the _De Animalibus ad Caesarem_, and that this source was an Arabic one. This point settled, it becomes possible to establish another. One of the copies of the _De Animalibus ad Caesarem_[86] has the following colophon: ‘Completus est liber Aristotelis de animalibus, translatus a magistro michaele in tollecto de arabico in latinum.’ Now if the version was made in Toledo, it was probably posterior in date to the _Physionomia_. This indeed is no more than might have been asserted on the ground of common likelihood; for, when a compilation and a complete version of one of the sources from which it was derived are both found passing under the name of the same author, it is but natural to suppose that the first was made before the other, and that in the interval the author had conceived the idea of producing in a fuller form a work he had already partially published.