170 Mr. R. H. M. Bosanquet on Temperament, 



from those of the diatonic scale, they will offend the ear. Helm- 

 holtz adduces, in support of this view, an experiment conducted 

 by him and Herr Joachim with the assistance of Helmholtz's 

 harmonium with pure scales. The result appeared to prove con- 

 clusively that the eminent violinist employed pure scales. And 

 it is sought to infer from this that the sequences of the diatonic 

 scale commend themselves naturally to the ear in all cases. This 

 is more than can be legitimately deduced from the experiment. 

 There is no doubt that good violinists play substantially perfect 

 concords, in general at least. It is clear that they cannot do so 

 in all cases ; for there are, as I have said, many passages in or- 

 dinary music which do not admit of being constructed with 

 perfect concords. And the artist who habitually plays perfect 

 concords (when away from the piano), may very well play 

 single notes in the same way. But there is no doubt, in the 

 first place, that all violinists play differently when accompanying 

 a keyed instrument ; and in the second place, the experiments 

 of Cornu and Mercadier, of which Mr. Ellis has given an 

 account, showed that the artists, whose performance was there 

 examined, generally played their thirds very sharp when freed 

 from accompaniment, though they played them true when the 

 harmony required it. 



But this seems to me eminently a question in which progress 

 may be made. It is commonly treated as if it were impossible 

 to bring any observation of one's own to bear upon the matter ; 

 but I have come to the conclusion that a process of education, 

 such as my own ears have gone through, and constantly repeated 

 experiments, such as those I have been in the habit of making 

 with the assistance of persons possessing ears of the most excep- 

 tional delicacy, are necessary to enable an opinion to be formed 

 on the matter. I do not think that any single experiments can 

 compare as evidence with such as have been repeated day after 

 day for a long period, until the effects to be observed have be- 

 come perfectly familiar. 



First, as to diatonic scales. The most striking effect of the 

 diatonic scale is, to my ear, the difference between the major and 

 minor tones : this difference is a comma ; and I now perceive it 

 with perfect clearness. The impression it makes varies accord- 

 ing to whether I have been using the diatonic scale much or not. 

 If I come back to the instrument (the 53 harmonium, which has 

 diatonic scales within a very close approximation) after a consi- 

 derable absence, and especially if I have been hearing much or- 

 dinary music, the impression made by the unequal tones is dis- 

 agreeable. When I first realized the difference I thought it 

 horrible. Few, except highly gifted musicians, can realize this 

 difference without frequent repetition of the experiments; but 



