560 PHYSIOLOGY 



In ordinary respiration the stream of air never passes higher 

 than the anterior inferior border of the superior turbinate bone, so 

 that it does not come in contact with the olfactory mucous membrane. 

 The sensations of smell which are aroused during ordinary respiration 

 depend on diffusion from the respiratory air into the still air of the 

 upper olfactory portion of the nasal cavity. The direction of olfactoiy 

 attention is achieved by sniffing ; in this act the nostrils are dilated 

 and the direction of the anterior part of the nasal respiratory chamber 

 altered, so that the stream of entering air is directed towards the 

 upper olfactory portion of the cavity. 



The fact that we are able to perceive smells when breathing 

 normally shows that the odorous substance must be diffusible, i.e. 

 gaseous in form. The amount of substance necessary to excite sensa- 

 tion is extremely minute. Thus '01 mg. of mercaptan diffused in 

 230 cubic metres of air is still distinctly perceptible. In this case a 

 litre of air would contain only '00000004 mg. of the substance, and 

 the amount actually in contact with the olfactory epithelium would 

 be still smaller. It is possible, however, to show the presence of these 

 odorous substances in air by physical means. Tyndall pointed out 

 that air containing a small proportion of odorous substances absorbed 

 radiant heat to a much greater degree than did pure air. Thus in 

 one experiment air containing patchouli absorbed radiant heat thirty- 

 two times as strongly as the pure air. Most odorous substances possess 

 large molecules and have therefore high vapour densities. On this 

 account the smell tends to hang about objects, the rate of diffusion 

 of the vapour being only small. 



Since the endings of the olfactory cells are bathed in fluid, it is 

 evident that the odorous substances must be dissolved by this fluid 

 before they can excite the olfactory nerve fibres, and in the case of 

 aquatic animals we know that the projected chemical sense, which we 

 call smell, can only be aroused by substances in solution. It is 

 difficult to show in man that the nerve-endings can be excited by 

 solutions. Most of the experiments have been made with solutions 

 which had an injurious effect upon the olfactory epithelium. Accord- 

 ing to Aronsohn it is possible to excite sensations of smell if the nasal 

 cavity be filled with normal saline fluid, containing a very small propor- 

 tion of the odorous substance. To this experiment it has been objected 

 that it is almost impossible to fill the nasal cavities without leaving 

 some air spaces so that the olfactory sensation obtained might have 

 been due to stimulation of the olfactory cells in such a space. There 

 is, however, no a priori reason to deny the probability of Aronsohn 's 

 conclusions. 



Many olfactory stimuli owe their peculiar character to the simul- 

 taneous stimulation of other kinds of nerve-endings. Thus a pungent 



