the constant value .0004 inch, an error of one line in the measured coincidence corresponds to an actual error of sixty-five ten millionths of an inch, counting from the first line.

The following is a description of the ruled bar to which the test here described has been applied.

This bar was ruled for the use of the American Watch Co., upon a machine built for me by Mr. Chas. V. Woerd, the mechanical superintendent of the watch factory. By his kindness I am permitted to exhibit it at this meeting.

On Tuesday, Aug. 9, commencing at 9h. 10m. A. M., Mr. Ballou subdivided four inches into forty equal parts. At 9h. 40m. A. M. the machine was started at 2500 spaces to the inch, and at 11h. 20m. P. M. the space of four inches had been subdivided into 10000 equal parts. The ruling carriage was then set back for coincidence with. the first line and the machine was started at 1000 spaces to the centimetre. At 1h. Om. P. M. of Aug. 10, the space of one decimetre had been subdivided into 10000 equal parts. The carriage was then set back for coincidence with the first line and the decimetre space was again subdivided into 100 equal parts.

It was hardly to be expected that in the continuous work of the machine for more than twenty-nine consecutive hours the theoret ical values of the coincidences should be maintained under the wide variation of temperature which took place between the day and the night. Moreover the tremors occasioned by the heavy machinery of the factory, and especially the blows of a power hammer which is not very far distant, were sufficient to account for a portion of the errors introduced. Nevertheless the matching of the (triple) lines representing of an inch with the corresponding lines of the band is nearly perfect. In the metric band. there is at one point a deviation amounting to about one-half of the width of the lines, but the coincidence is soon recovered.

In counting the coincidences between the two bands, of which there are 158, the following method was pursued. After Mr. Ballou had completed the graduation, he arranged a simple carriage for facilitating the count. He carefully counted the coincidences and communicated the results to me. I then compared them with the computed numbers.

It is to be noted especially that Mr. Ballou had no means of knowing whether his count would agree with the actual number of

lines forming a coincidence. The deviations from the computed values are given in the following table, for each of the 158 coincidences.

It will be seen that the greatest cumulative error is near line 2000, the maximum value of the difference between the two values of the two units at this point amounting to about one thirty-thousandth of an inch. It is to be noted that the apparent periodicity of the numbers representing the coincidences does not necessarily indicate a periodicity in the screw. In fact, this method of comparison gives us no information on this point, since the coinciding lines in the two systems were ruled at the same part of the screw. Finally, the four inches of this bar are nearly standard at 62°, and the decimetre is nearly standard at the same temperature.

FREEZING MICROTOME. BY THOMAS TAYLOR, of Washington, D. C. ALL microscopists who take an interest in the study of histology and pathology have long felt the necessity for a better method of freezing animal and vegetable tissue than has heretofore been at their command. The methods of hardening tissue by chemical agents need not be described at this time, as they are well understood, and moreover are successful as far as they go. The principal objections to the whole system are that the tissues are more or less distorted by the chemical solutions used to harden them; the chemicals are expensive, and the tissue requires to be exposed to their action for a considerable time, in some cases for a period of months. To avoid these objections the freezing microtome is introduced. Ether and rhigolene have been employed with some degree of success, but great disadvantage attends the use of vapor, or spray of these liquids. Both are expensive and generally very unsatisfactory in action; they cannot be used in the presence of artificial light because of danger of explosion. Two persons are required to attend to the manipulations, one to force the vapor into the freezing box, while the other uses the knife. The moment the pumping of the ether or rhigolene ceases, the tissue operated on ceases to be frozen, so ephemeral is the degree of cold obtained by these means. The Rutherford" machine is also defective. In using it, I with others have found it necessary to immerse the whole machine in a freezing mixture in order to get a sufficiently low temperature to bear on the tissues. The microtome which I have constructed presents all the advantages of any plan hitherto employed in hardening animal or vegetable tissues for section cutting, while it has many advantages over all other devices employed for the same purpose. The method of freezing the tissues which I propose may, when compared with any other mode heretofore introduced, be said to be instantaneous. So quickly is the gumwater frozen, which is used to secure the adherence of the tissue on the freezing chamber, that it is found necessary to place the gum and tissue in position before the freezing process is commenced, otherwise the gum would be so firmly frozen that the tissue would remain loose on the frozen gum surface.

The general appearance of my freezing apparatus resembles that of an ether microtome; but in their working parts and results, the two instruments differ materially. The method of using my microtome can be best explained in connection with the illustration.

Fig. 1 represents the microtome when secured to a table. a is the freezing chamber; it resembles a pill box, constructed of metal and may be of any dimension required, provided it is in harmony with the framework which holds it in position.

A brass tube enters it on the right side as shown. On the left side is a second tube similar to the other. The first tube is used to fill the box a with a freezing liquid, salt and water, of the temFIC.I.

perature of zero. The object of the second is to remove the air from the box as the liquid flows into it, and also to furnish an exit for the latter. This tube is bent at right angles, its inner portion ascending vertically to within an eighth of an inch of the top of the chamber (a), thus bringing the freezing mixture directly in contact with the top of the box. In order to secure a constant current of freezing liquid, I place above the microtome on a bracket a pail filled with finely chipped ice, combined with an equal quantity of common salt. The ice and salt quickly melts, the liquid thus formed consisting of salt and water is made to flow through a rubber pipe attached to the lower part of the pail, while the other end of the same pipe communicates with the right hand tube of the chamber at a. A rubber pipe is likewise secured to the second brass tube described. The water should not be allowed to waste. I secure it in a second pail to be returned to the first pail for continued use. In this way great economy is secured in the freezing material. I find it necessary as a matter of further economy, to limit the rate of exit of the salt and water. I therefore

make the supply pipe somewhat larger than the exit pipe, and to economize still further, I secure within the lower end of the exit pipe a glass tube the outer diameter of which is about the thirtysecond part of an inch. Sometimes in practice it is found that a small particle of vegetable matter will close up the orifice of the glass tube and prevent the constant flow of water. By means of a pin this may be removed and the flow at once established. Should a stoppage of flow arise from any other cause, the glass tube may be removed from the rubber pipe and blown through from its smaller end, which will remove all obstructions. With a little practice the freezing process may be constantly kept up as long as desired, by simply renewing the chipped ice and salt in the upper pail.

In all my experiments I prefer a long knife such as is generally supplied for section cutting. Should the cold within the chamber (a) be too intense, the edge of the knife is liable to be turned and the cutting will be imperfect. When this occurs I stop the flow of water through the chamber by placing a spring clothes-pin as a clip on the upper tube. By the use of the freezing microtome, the softest tissue may be hardened and cut in sections, and mounted within a few moments, all chemical hardening processes being rendered wholly unnecessary. a, fig. 2, represents a sectional view of freezing chamber, as seen apart from its framework. b contains a solid cylinder of box wood to which the frozen chamber is secured. The arrow represents the inward and outward flow of water. At the bottom of the figure is the index screw by which the thickness of the section is regulated.