84 C. J. MERFIELD. 
These elements agreed fairly well with those computed by Dr. 
Knopf, and published in the Astronomische Nachrichten 3394, 
that is if ten minutes of arc be applied by addition to the longi- 
tude of the ascending node, as there published. The computation 
of the co-ordinates of the comet showed that this correction was 
necessary to make the given figures consistent, the error being 
evidently a typographical one. 
The elements agreeing so well, the author decided to adopt the 
ephemeris, computed by him from Dr. Knopf’s elements, in com- 
paring the various observations for the purpose of obtaining normal 
\ places. 
The first normal place was constructed by comparing with 
_the computed ephemeris ten observations from several European 
observatories, taken between the dates 1896 November 26, and 
1896 December 4. These observations were culled from the 
Astronomische Nachrichten, Comptes Rendus de l Academie, and 
the notes of the Royal Astronomical Society. 
Consulting the table, denoted by the Roman numeral (I.), there 
will be found a complete list of observatories, the observations of 
which have been employed in constructing the first normal. After 
reducing the times to the meridian of Greenwich, they have been 
corrected for aberration of light, and tabulated in column three 
as the day and fraction of the year 1896. The residuals, after 
comparing the observed co-ordinates with the computed ephemeris, 
are given in columns four and five. 
The second normal place was computed from observations, — 
taken by Mr. Tebbutt, between the dates 1897 March 5 and 1897 
March 15, some eight observations being compared in a similar 
manner as in the construction of the first normal. (See Table IL.) 
In the construction of these normal places, the means of the 
residuals have been applied to the right ascension («), and the 
declination (5), taken from the computed ephemeris corresponding — 
to the mean of the times. 
