Having thus disposed of all questions concerning the horse's apparent feats of reason and memory, let us turn to those in the field of sensation. We shall begin with vision. That Hans was unable to select colored pieces of cloth merely upon the basis of color quality, without reference to their order, was shown in Chapter II. It would, however, be somewhat hasty to infer color-blindness from this fact, as did Romanes[32] on the basis of similar unsucessful responses on the part of a chimpanzee ("Sally" of the London Zooelogical Garden). It is much easier to explain the failure of the horse than that of the monkey on the basis of intellectual poverty, a poverty of associative activity. It presumably can discriminate between the various colors, but it cannot associate with these their names. The existence of chromatic vision in the lower forms is by no means as unquestionable as is assumed by popular thought. Even teleological considerations which are often brought forward (especially that of the ornamental and protective coloring of so many animals) can never do more than establish a certain probability. For definite proof, we need data given by observation (we have none in this case), or experimental evidence. Such evidence we have, but it is insufficient in quantity and unfortunately most of it was obtained under inadequate experimental conditions.[W] We know nothing regarding chromatic vision in the horse, though we have often had trained horses which apparently possessed color discrimination. The earliest report of this kind I find in a work published in the year 1573.[36] Here we read that a number of Germans exhibited two horses in Rome which could, upon request of their masters, point out those persons among the spectators who were wearing stockings of any designated color. The passage, "conoscevano i colori", (they recognized the colors,) proves nothing and no one has ever heard, even in modern times, of a horse that actually knew colors.