CULTURE OF WHEAT WITHIN THE TROPICS. 
119 
by no means of such a quality as the results of the trials in the two last islands 
had led me to expect, the grain being inferior, in the number contained in the ear, 
as well as in bulk and weight, to the original seed; but in the absence of all 
details, no opinion can be formed as to the causes of this degeneracy. 
In point of climate, that of New-Providence, notwithstanding its being nine 
degrees removed from the line, cannot, in any very material degree, differ from 
that of Antigua, since, if we assume, with Humboldt, the mean temperature of 
October, as representing with tolerable accuracy that of the entire year, we shall 
have from 78 to 79 for the mean annual temperature of New-Providence ; which 
differs by from two to three degrees only from the mean temperature of October, 
for the latitude of 17. Mr. Butler, in the notes of his experiment, states that 
the temperature, in the month of February, the first of the growth of the Wheat, 
descended as low as 59, which, at that early stage, could have had little more 
effect than to retard its progress some few days, since we find the whole of the 
quantity planted did not make its appearance above ground before the 18th, or on 
the 19th day after it was put into the ground; while the ears did not commence 
forming for 37 days more, being 55 from the day of sowing; in which, and the 
succeeding month, the temperature ranged from 70 to 80,* to which months, for 
the latitude of Antigua, Kirwan assigns the mean temperatures of 78.5 and 
794. Hence we can hardly attribute any great influence to the effect of climate, 
since, during the period occupied in the formation of the ears and the filling of 
the grain, the extreme difference of temperature could hardly have amounted to 
five degrees. The season of Mr. Butler's experiment, as we learn from his 
report, was by no means favourable, being very dry, while the peculiar circum¬ 
stances of the Bahama islands put artificial irrigation wholly out of the question. 
The soil we may presume to have been similar to that described by Mr. Lees as 
the subject of Mr. Storr’s experiments, consisting of “pulverized Madrepore 
rock, and minute fragments of sea shells this, however, was improved by Mr. 
Butler, by a dressing of good Sheep-pen manure, mixed with calcareous sand, a 
* We shall presently see, from Mr. Gilbert’s table of the mean temperature of each of the 
four months occupied with his experiments (and which, with the exception of August, are the 
hottest of the year), that this range varies but little from the mean temperature of those hot 
months in Antigua, which from that table appears to be 77.7, while the mean of Mr. Butler’s 
extremes gives 75. The mean annual temperature of the Antilles, according to Humboldt, is 
81.5; of Cumana (lat. 10° 27. 52.), 81.26. and at Vera-Cruz and the Havanna, 78.08. The follow, 
ing table exhibits, according to the same author, the mean temperature of the two coldest and two 
hottest months: 
Coldest. 
Hottest. 
Diff. bet. hot¬ 
Dec. 
Jan. 
July 
Aug. 
test & coldest. 
Havanna, lat. 23° 8. 
71.78 
70.16 
83.11 
83.54 
13.38 
