222 Selenography : its Past, Present, and Future. {April, 
it would add comparatively little to the work, though much to 
the accuracy of the results, were the investigation made to 
depend on three separate series of objects on different sides 
of the lunar equator, as Poisson has shown; whilst by 
mutual comparison some of the incidental errors could be 
eliminated. By means of a series of simultaneous measures 
from the central mountain of Manilius, a series would be 
obtained of like nature to the former, but free from the 
effects of the real libration of the moon and the irradiation 
at the limb, which would of course be also eliminated 
as much as possible from the former series by measures 
being taken from all four directions. By the mutual com- 
parison of the two series, the accuracy of the results of the 
discussion of the first would be improved, whilst means 
would be afforded of comparing with the results of Bouvard 
and Nicollet. The comparison with the investigation of 
Wichmann would be independently done, as the crater 
Mosting A would be embraced in both series. Employing 
suitable instrumental means, the errors due to mere instru- 
mental effects might be reduced considerably from those of 
Bouvard, and rendered equal, perhaps, to Wichmann, and 
from a series of from two to four hundred measures a 
trustworthy conclusion with regard to the real lunar libration 
might be arrived at. 
Attention has now been directed, not only to the present 
position of selenography, but also to the progressive stages 
by which it has arrived at the condition it now stands in— 
a little known history; whilst the direction where further 
selenographical work appears most requisite, and which 
should be the first to be supplied, have also been indicated 
in some detail, and our present knowledge on these points 
described. The necessity of the introduction into lunar 
observations of micrometrical measures has been again 
urged with special stress, for it is only on a foundation of 
exact micrometrical work that the desirable accuracy in the 
delineation of the lunar formations can be obtained; for, 
though practice renders the results of an experienced 
observer sufficiently faithful, as a rule, for many purposes, 
yet in the comparison of drawings of the same formation at 
different epochs, all certainty is lost with regard to differ- 
ences, unless based on adequate measures, as is exemplified 
by the results of Linne, and still more so in the case of the 
twin Pytheas of Mayer, to say nothing of minor instances. 
In all branches of science too much faith is put in the 
results to be achieved by mere piling up of observation on 
observation, and trusting, by so doing, some fortunate 
