skip to content

Darwin Correspondence Project

From Arthur Nicols   [before 20 March 1873]1

I do not know whether you will consider this a crucial experiment as to the comparative acuteness of the sense of smell in dogs and cats, but perhaps it may be useful in adding one fact to the discussion on the part played by this sense in guiding animals home.2 I have long had reason to think that the sense of smell in cats is much less highly developed than in dogs and even many other animals, because, among other things, we see the difficulty cats often seem to experience in finding food thrown down to them, unless they see it fall, bobbing their noses about on the floor in search of it, even when it is no distance from them. A few days ago, therefore, I prepared some dozen or so of dainty pieces of meat, both raw and cooked, and some pieces of fried cod and herring, and, taking my dog into a room from which every ray of light had been excluded, threw pieces of the meat into different parts of the room. As might have been expected, each piece was found by him almost as soon as the first could be eaten. The house cat was afterwards tried in the same room, and had great difficulty in finding pieces dropped close to her, failing altogether in securing some of them. What the dog accomplished in the space of a minute, the cat could not do in a quarter of an hour; for, on letting light into the room, I found pieces of the fish lying about in the further corners. There was no comparison between the one and the other in the manner of searching for the food. The dog went to work with confidence, and, after a few seconds employed in sniffing round, could be heard eating until every piece of meat had been found. The cat, on the contrary, walked about mewing, and seemed to have no idea of the presence of the fish until she was close to it. The cat was quite familiar with me, and had been kept a long time without food intentionally. I used fish because it was a food to which she was accustomed, and calculated to emit sufficient smell. The result impressed me with the conviction that cats discover food by smell with very indifferent success; whence perhaps it may be inferred that their perceptions generally through this sense are more feeble than those of some animals.


The date is established by the relationship between this letter and the letter to Arthur Nicols, [20 March 1873].
Alfred Russel Wallace had suggested in a letter to Nature, 20 February 1873, p. 303, that the homing ability displayed by some animals could be explained by an acute sense of smell; Nicols had criticised the idea in his letter to CD of 21 February 1873.


Compares sense of smell in dogs and cats.

Letter details

Letter no.
Robert Arthur (Arthur) Nicols
Charles Robert Darwin
Source of text
Nicols 1885, pp. 51–2

Please cite as

Darwin Correspondence Project, “Letter no. 8817F,” accessed on 18 September 2019,

Also published in The Correspondence of Charles Darwin, vol. 21