The typical reflex is a much simpler act than the typical instinct, but it is impossible to separate the two classes on this basis. At the best, this would be a difference of degree and not of kind. Among reflexes, some are simpler than others, but even the simplest is compound in the sense of being a coördinated movement. The knee jerk is simpler than the flexion reflex, and this is simpler than the scratch {108} reflex, which consists of a rapid alternation of flexion and extension by one leg, while the other is stiffly extended and supports the trunk. Coughing, which would be called a reflex rather than an instinct, consists of a similar alternation of inspiration and forced expiration, and swallowing consists of a series of tongue, throat and gullet movements. These compound reflexes show that we cannot accept the simple definition that is sometimes given for an instinct, that it is a compound of reflexes. Such a definition would place coughing and swallowing among the instincts, and so do violence to the ordinary use of the word. In point of complexity, we find a graded series ranging from the pupillary reflex at one extreme to the nesting or mating instinct at the other, and no sharp line can be drawn on this score between the reflexes and the instincts. Another distinction has been attempted on the basis of consciousness. Typically, it may be said, a reflex works automatically and unconsciously, while an instinct is consciously impulsive. The reflex, accordingly, would be an unconscious reaction, the instinct a conscious reaction. But this distinction also breaks down on examination of cases. The pupillary reflex, to be sure, is entirely unconscious. But the flexion reflex is a little different. When unimpeded, it occurs so promptly that we are scarcely aware of the painful stimulus before the reaction has occurred. But let the reaction be hindered--either voluntarily or, for instance, by the foot being seized and held--and a strong conscious impulse is felt to pull the leg away; so that here the flexion reflex would belong among the instincts, according to the proposed distinction. Similar remarks would apply equally well to coughing, since a strong impulse to cough is felt if the coughing movement is checked. Sneezing, a protective reflex, is usually a slow reaction, giving time for a conscious impulse to {109} sneeze before the reaction takes place. The same is true of scratching and of swallowing, and of a number of other reflexes. In short, it is impossible to draw a satisfactory line between reflexes and instincts on the basis of conscious impulse.