110 Dr. Hale on the Meteorological Journal 
set; andthe fourth was uniformly at ten in the evening. It is 
much to be regretted that the observation at sunrise was not 
continued a sufficient length of time, and with sufficient regu- 
larity, to be of much value. 
Abstracts of the thermometrical journal for thirty-three years 
have already been published in the Academy’s Memoirs; the 
first seven years by Dr. Holyoke himself in 1793, vol. Il. p. 89., 
and the next twenty-six years prepared by Mr. Clapp in 1818, 
vol. [V. p. 361. But an abstract of a journal of this kind can 
be, only to a very small extent, a substitute for the journal in 
detail, since the purposes for which different persons consult it 
are as different as the objects and pursuits of the several indi- 
viduals; while the abstract can only have reference to the 
objects and views of the author of the abstract. 
As an indication of the mean temperature of our climate, this 
journal will not perhaps be regarded as very conclusive. Indeed 
this is true of nearly or quite all the numerous observations 
which have been made with the thermometer for the same pur- 
pose in different parts of the world. Although this instrument has 
now been carried to such perfection, as to leave little more to 
desire, in regard to the instrument itself, as a measure of heat 
within the range to which it is applicable; yet when it is applied 
to the observation of the temperature of the atmosphere, it is 
liable to be influenced by so many circumstances, which have 
not hitherto received their due share of attention, that we cannot 
yet place entire confidence in the results. 
It is not easy to fix upon such hours of observation, as will 
give a true mean of the heat of the whole day. ‘The surest 
method would be io take the extremes of cold and heat for each 
day. But self-registering thermometers, by which alone such 
