It was not a common thing in those days for young girls to study Greek and at the time of graduation she received a letter from the principal expressing in the warmest terms his appreciation of her Greek and of her character.

Will you have the kindness to engage Miss Saunders as a computer at the Observatory for one year at a salary of $600 a year.

The first women who joined the Harvard Observatory staff in 1875 ... spent their days conducting mathematical calculations for the male astronomers who worked at the telescopes overnight. The women computers took the astronomers' observation notebooks and "reduced" the data recorded inside - averaging numbers, and correcting them for refraction, parallax, and error inherent in different instruments in order to record an object's absolute position in the sky.

... was a specialised library, with rows of books that contained the tables and formulas [the computers] needed to perform their calculations.

... was often involved in meridian circle computations, which were for many years astronomy's most reliable source of positional information about stars. Meridian circles were telescopes which were locked into a position that could scan along the meridian (the great circle that runs through the celestial north and south poles, and the local zenith). Rather than chasing after stars, these telescopes waited for stars to pass over them, and took down information about the timing of that procession and the angles to the observed star. Taking that information and changing it into stellar positions was an intense process that Winlock mastered and employed over the course of nearly three decades of work, while simultaneously overseeing the thirty-eight-volume 'Observatory Annals', which collected tables of variable star positions and observations within star clusters.

Note by William A Rogers. - My connection with this work is limited to the methods of discussion adopted, and to an examination of the numerical results obtained. Beyond this, all the work in the preparation of this paper has been done by my assistant, Miss Winlock, and she is entitled to all the credit therefor.

It is the purpose of the present paper to discuss the modern observations of such polar stars north of +7º declination as are found in the Harvard College Catalogue of 1213 Stars. Of the 130 stars in this list, 68 are found in the Fundamental Catalogue of Dr Auwers.

There will be no attempt to determine the proper motions of these stars, but the places determined for the epoch 1875.0 will serve a useful purpose in future discussions of this element. ... .

All the observations employed in this discussion will be reduced to the System of Publication XIV, of the 'Astronomische Gesellschaft', either directly through the medium of the fundamental stars common to the observed and the fundamental systems, or indirectly through the medium of the Harvard College Catalogue.

For many reasons, it was found advisable to construct a yearly ephemeris of each of the stars in the proposed list, extending from 1860 to 1885. For the fundamental stars of the list, the places for 1880 and for subsequent years were taken directly either from the catalogues of the 'Astronomische Gesellschaft' or from the 'Berliner Jahrbuch'. For the years 1871 to 1879 inclusive, they were obtained by applying to the yearly ephemerides of the Gesellschaft the corresponding corrections by which the provisional system is reduced to the system of Publication XIV.

For the places of the fundamental stars between 1860 and 1870, and for the places of all non-fundamental stars for the entire period between 1860 and 1885, the reduction-elements given in the Harvard College Catalogue were employed.

For stars below 85° north declination the development of $\alpha$ and $\delta$ in terms of the first, second, and third powers of the time will be sufficiently accurate for the limit of fifteen years. For the reduction of stars near the pole, the problem becomes more difficult. Since the method of development by differential coefficients in terms of the ascending powers of the time has necessary limitations in its application, it has been thought advisable to give an illustration of the various methods by which the coordinates for any time $t_{0}$ are reduced to those for any time $t'$. The star Groombridge 1119 is selected for this purpose. The reductions for precession and for proper motion will be considered independently.
...
The second part of this paper will be comprised under the following subdivisions:-

(a) Treatment of the proper motion for close polar stars.

(b) Yearly ephemerides of all stars within 3° of the pole, between the limits 1860 and 1885.

(c) Tabular values of the terms $U^{I}, U^{II}, U^{III}, U^{IV}, U^{V}$, &c., carried as far as will be necessary to give the exact reduction for 40 years.

(d) Tabular values of the proper motions at intervals of 8 years for close polar stars, and at intervals of 20 years for all other stars.

(e) Data for the reduction of the different catalogues employed to the system of Publication XIV.

(f) Tabular values of the systematic relations between the catalogue of Publication XIV and the different catalogues compared.

(g) Final catalogues of 130 stars resulting from this discussion.

(h) Comparison of the final catalogue with the various catalogues from which it has been derived.

The review of the paper in the Sidereal Messenger ends as follows [1]:-

To show how the work was done, the necessary formulae are stated in order fully, and the tables of constants given, and then a single star Groombridge 1119, for epoch 1875.0, is taken as an example and its reduction shown in detail. The method of the paper seems excellent and complete, and is a credit to its authors. We shall look with interest for succeeding papers, in which will appear the discussion of other kindred topics named at the close of the one before us.

The results were published in phases between 1891-1896, and Winlock's dedication to the project allowed Harvard to be one of the first of the collaborating observatories to publish its portion of the project.

From an examination of a plate taken on August 14, 1901, with the Bruce Telescope, Dr Stewart found an asteroid having the great southern declination, -62°. As no known asteroid was so far south at that time, a series of photographs was taken, from which, and from one previous photograph, the approximate positions ... have been derived. A circular orbit was first computed by Miss Anna Winlock. This gave the surprising result that the heliocentric diurnal motion exceeded 2200", corresponding to a distance from the Sun less than that of any known asteroid. As Professor Newcomb was spending a few days in Boston, he courteously undertook, with the assistance of Miss Winlock, to determine elliptical elements for this object. ... From these elements it appears that the great peculiarity of this orbit is the ellipticity, which exceeds that of any known asteroid. ... At the time of discovery the asteroid was near perihelion, and therefore was moving very rapidly around the Sun, at a distance of about 1.6. An approximate ephemeris for Greenwich Midnight ... shows that the asteroid is moving rapidly north, and is now within reach of the telescopes in Europe and the United States.

To friends and kindred she leaves a rich legacy simply by having lived. Her delicate sense of humour gave a very human touch to a nature too spiritual and too intellectual to be understood by all. Her gentle loving kindness veiled in part the power of her intellect. Doubtless many knew her without realising how far beyond the average were her powers of mind. She seemed not to realise it herself. For a little thing that others did she was prodigal with hearty praise while she quietly, but persistently ignored real achievements of her own.

Miss Winlock was the eldest daughter of Professor Winlock, the former director of the Observatory, and began her astronomical work soon after the death of her father in 1875. She assisted the late Professor W A Rogers in the reduction of his meridian circle observations, including those of the zone 49º 50' and 55º 10' north declinations, and after his death she took charge of the reduction of those still remaining unpublished. In addition to this work, she undertook many other contributions of importance, some of them belonging to the department of theoretical astronomy. Her care, efficiency, and good judgement were of great value to the Observatory.