When we leave these simple modes of feeling, and consider the affections of the higher senses of hearing and sight, we are confronted with a new problem. Sensations of the first class are almost purely homogeneous, they owe little or nothing to memory and imagination. Moreover, though generally due to an external cause, they are in the case of smell or taste occasionally, and in that of pain frequently, excited by causes within the organism. It is not, therefore, a matter calling for comment that in such cases the transferred idea should assume a definitely sensory form. But when the organs of sight or hearing are sensibly affected, past experience has taught us to look for an external cause; the line between _idea_ and _sensation_ is here sharply drawn and clearly understood.[87] The line, indeed, as drawn by common use may not correspond to any real distinction in the nature of the experience itself. Ideas may be only paler sensations, and a train of thought nothing else than a series of suppressed hallucinations. But at any rate the distinction, whether fundamental or not, serves a useful purpose as a rough-and-ready means of classing our mental experiences. A visual or auditory image either is on the same level of intensity as the series of impressions which represent for us the external world, or it falls below that level. In the former case we call it a sensation or percept, in the latter, an idea. Sensations and percepts may be again subdivided, as objective or hallucinatory, according as they do or do not correspond to a supposed material cause. In the experiments described in the first five chapters, it will have been observed that when the transferred impression was of a visual nature it generally remained ideal, rising occasionally, however, as in some of the experiments with hypnotised percipients, and in Mr. Kirk's cases, to the level of a complete sensory hallucination or quasi-percept. In the present chapter it is proposed to deal with auditory and visual phantasms which, so far as can be judged, were of an ideal kind, though one or two of the cases cited may seem to approximate to sensory embodiment. The more striking hallucinatory effects will be reserved for later chapters. _Transference of Ideas._