OF NEBULAE AND CLUSTERS OF STARS. 
5 
recorded, so that any suspiciously large deviation from the mean of all may be at once 
noticed and traced to its origin in the sweeping books. My reduction was of course 
based on the means of all these (rejecting such as were obviously and grossly faulty), 
and might therefore, pro tanto, be regarded as of superior authority. This consideration, 
joined to that before adduced, decided me to retain those places in the present Catalogue 
which had been derived from this source, except in a few instances (specified in the 
notes) when it proved, by careful examination of the causes of discordance, that actual 
mistakes had been committed. And I must not omit to add that the comparison so 
instituted with M. Auwers’s results has led me to the detection of several grave errors 
in my own work which would certainly have otherwise escaped notice (and in some 
cases have caused the loss of future observations by missetting the telescope), and whose 
rectification has added materially to its value. On the other hand, as no human work 
is perfect, I have been led to notice some errors in M. Auwers’s work itself, which are 
set down in a list of errata and corrigenda at the end of this Catalogue ; and besides, a 
good many cases in which, owing to mistakes in the printed catalogue in the volumes of 
the Philosophical Transactions (many of which stood corrected in MS. in the margin of 
the copy of those Transactions in my possession, and many more have been silently 
detected and rectified by Miss C. H. in her subsequent computations), his calculations 
have been founded on erroneous data, and have therefore led him to assign erroneous 
places to the objects so affected. Thus on every account the result has been what may 
be considered a complete expurgation of both our catalogues. 
It remains for me to say a few words on the way in which the reduction to 1860 and 
the calculation of the precessions have been performed by Mr. Kerschner, the com- 
putist employed by the Astronomer Royal for that purpose. The whole work has been 
executed on printed forms, which being preserved may at any time be referred to. 
Since error in computation, however practised the computer, and however checked, is 
always possible, and occasional error of copying, especially when the order of the entries 
has to be rearranged, is absolutely unavoidable, I considered it incumbent on me to 
recalculate, seriatim from my original MS. Catalogue for 1830, and taking for granted 
the precessions set down in the fair copy, for 1880, the places both in R.A. and P.D. of 
every object included in the Catalogue ; keeping an eye meanwhile to the precessions 
themselves, and their signs, to seize the least indication of error in that quarter. It 
would have been too laborious to recompute these. As for the precessions in P.D., 
their regular progression of itself ensures their correctness, as far at least as the integer 
seconds and the first decimal place. A pretty considerable number of errors (most of 
them of little moment) was thus detected and corrected — not more, however, than might 
reasonably be expected in the work of the most expert computisk in so extensive a work, 
consisting of between nine and ten thousand computed entries (taking both elements), 
and traceable moreover in many instances to obvious misreading, and in some to actual 
misentry on my part, of figures in the original MS., which but for this further examina- 
tion would also have escaped notice altogether. 
The correction of these and the other errors already spoken of necessitated, in a great 
