ON THE ANATOMY OF FISHES. 
281 
pressure on the somewhat improbable supposition that the Fish remained at exactly 
the same depth during the whole time that the barometrical variation was in progress. 
The maximum range of variation in atmospheric pressure, as measured by the 
barometer, is about 2 inches, but such variations occur only at considerable intervals 
of time, and then may take many hours to accomplish. But even in this extreme 
case the atmospheric pressure variation might be negatived as far as the Fish was 
concerned by ascent or descent in the water to the extent of 27 inches, or more or 
less completely masked by similar movements of less extent during ordinary loco- 
motion, or by the rise and fall of the tide in the case of a few marine or estuarine 
Siluridse. 
We have not been able to examine any records relating to barometrical variations in 
regions where the Siluridae or Characinidae are most abundant, but the extent to which 
the Ostariophyseae of the British Isles are exposed to atmospheric pressure variations 
may be readily tested. The only Ostariophyseae inhabiting Great Britain are the 
Cyprinidae, the various species of which are among the commonest and most abundant 
of British Fishes of fresh- water habitat. Ten genera and fifteen species of Cyprinidae 
are known to inhabit this country (10a), and, omitting certain genera^^ which have 
small bone-encapsuled air-bladders, there will still remain eight genera and thirteen 
species with normal, well-developed air-bladders in connection with a Weberian 
apparatus. It may, therefore, be worth while to ascertain the extent and variety of 
the barometrical changes to which these Fishes are liable. With this object we 
carefully examined the variations of the barometer as registered by the barograph 
of the Birmingham and Midland Institute Observatory during the three years 
1886-1888. These records demonstrate (1) that variations of *50 inch and upwards 
very rarely occur in less time than 10 hours, and generally require from 12 to 
24 hours for their completion. In the three years on only two occasions did a 
variation so great as ‘50 inch occur in less time than 10 hours; (2) variations 
amounting to *50 inch and upwards in 24 hours are not common, occurring only 
fifteen times in 1886, fourteen in 1887, and seven times in 1888 ; (3) in only one 
instance did a barometrical rise or fall exceed 1 inch in 24 hours ; excluding this 
very exceptional case, the most striking variations within this period ranged from 
•60 inch to '82 inch, and required from 15 to 24 hours for their completion; such 
variations happened but seldom, and are recorded only nine times in 1886, four in 
1887, and four in 1888; (4) rapid and sudden variations, that is, those occurring 
within, say, 2-5 hours, are somewhat rare, and, almost invariably, are very small in 
amount, rarely exceeding ‘2 inch. It may be reasonably concluded, therefore, that, 
so far as the Cyprinidae of the British Isles are concerned, the variations of atmospheric 
pressure to which they are liable are too small in amount, too gradual in their pro- 
gress, and too infrequent to exert any appreciable influence upon an air-bladder 
already subjected to considerable hydrostatic pressure, especially as tlie latter, instead 
* Cohiiis and Nemacldlns. 
2 O 
MDCCCXCTII. — B. 
