TRANSACTIONS OF SECTION D. 1059 
weight to the albumen in solution failed in producing the slightest trace of 
oxidation, and that the ozonised air passed through the liquid quite unaltered. 
During the course of the experiments, and for six days following, the develop- 
ment of micro-organisms ceased; but at the end of that time, and notwith- 
standing the cotton wool plugs, the liquid became slightly turbid from the 
presence of organisms. 
As dilute hydrogen peroxide is without action upon albumen, the conclusion 
seems inevitable that albumen is practically indestructible by any atmospheric 
agency without previous splitting up by micro-organisms, and further that whilst 
micro-organisms cannot develop, and are probably killed in an ozonised atmosphere, 
their spores are not easily destroyed by its agency. 
These results confirm the surmise of the late Dr. Angus Smith, that putre- 
faction is a necessary preliminary to oxidation in all cases of natural river 
purification. 
4. A new Theory of the Sense of Taste. 
By Professor J. Berry Hayorarr. 
The author showed that ‘ quality’ in this sense depends upon the nature of the 
atoms found in the sapid molecule. A study of the periodic law demonstrates that 
similar tastes are produced by combinations which contain elements, such as 
lithium, sodium, and potassium, which show a periodic recurrence of ordinary 
physical properties. Among the carbon compounds, those which produce similar 
tastes are found to contain a common group of elements. Thus organic acids con- 
tain the group CO, OH ; the sweet substances CH,,OH. There is no relation be- 
tween quality of taste and gross molecular weight, except that substances either of 
very small or very great molecular weight are not tasted at all. 
SATURDAY, SEPTEMBER 12. 
The following Papers were read :— 
1. On a Model of the Whale. By Captain Gray. 
2. On the Hybridisation of Salmonide at Howietoun. By Francis Day, C.L.E. 
Among the papers on the fishes of the British Isles, which have from time to 
time appeared in the ‘ Proceedings of the British Association,’ are several respecting 
the Salmonide. Most of these communications have been restricted to investiga- 
tions into species and their distribution, or questions as to their artificial propaga- 
tion, but none to hybridisation. 
During the last eleven years, Sir J. R. Gibson-Maitland, at Howietoun, near 
Stirling, has devoted much attention to this subject, and gone to great expense in 
order to efficiently carry out the many experiments he has instituted, while he has 
likewise afforded me facilities for personally watching many of them, and furnished 
me with data as well as with specimens. This subject, which is among the most 
interesting problems of fish life, I have selected for communication to this meeting ; 
for, although only some conclusions have been arrived at, I cannot resist feeling 
that such important Scottish experiments ought to be recorded, so far as they 
have gone, in the ‘ Proceedings’ of this Association, now that itis holding a meeting 
in the country were they are being carried out. 
Willoghby and Ray, as early as 1686, alluded to hybrid fishes, and Pennant 
(edition 1812) observed ‘hybrid fish, for that such exist those persons who have 
paid most attention to the subject in ichthyology have not a doubt.’ 
When we consider that the ova of teleostean or bony fishes have, as a rule, to 
3x2 
