are the same as those already directed against printed history and examination cards-they go far beyond the routine required for the average case and they are likely to prevent expansion when this becomes necessary. The particular case, for example, may demand twenty urinalyses. to one hematologic examination, and yet a set form would allow about equal space for each.

Still another modification of the four-page form which has much in its favor is the use of special pages, of the same size (82x11), for the different laboratory examinations, such pages to be fastened within the booklet by metal clasps. These additional pages should follow, in general, the forms shown in Fig. 5 and not those of the type of Figs. 6 and 7. For easy identification, different colors of paper may be used for the dif ferent laboratory examinations, but, in order to prevent undue bulkiness, two, or even three, subjects may be included on a single page, somewhat after the plan suggested in Fig. 5. The number of the history and the name of the patient appears at the top of each laboratory page; and each laboratory phase-Urinalysis, Blood, etc.-is properly labeled.

The number of these colored laboratory forms which may seem de-. sirable will vary greatly; the physician in general practice will find two or possibly three such combination pages sufficient for all routine examinations. The specialist, on the contrary, will require forms suitable for particular purposes: field of vision, renal functional tests, diabetic charts, fractional gastric analyses and so forth; however, to attempt to reproduce, or even to discuss these various special forms is not within the scope of this work.

The letterhead size of page recommended in the various history forms and extra pages just described is preferable to other sizes not only because it affords sufficient space for nearly every purpose, but because correspondence relative to the case may be neatly and compactly filed away with it. The habit of stuffing cards and papers of various sizes in an envelope as a container for the records of each individual case detracts from system in the keeping of records. It is the author's custom to place five histories in a stout manila container, open at the top, and with a flap for the numbers of the records contained.

We come, finally, in the matter of materials, to methods of filing histories and of cross-indexing. Here again no attempt will be made to describe the many eminently satisfactory methods designed for this purpose. Any system which will readily enable the physician to find the particular history he wishes is satisfactory provided that, as the number of histories grows, the system does not become unwieldy. The alphabetical filing of records does become extremely unsatisfactory after a time, inasmuch as one finds it necessary to run through a certain number of histories, no matter how well subdivided the alphabet may be.

Distinctly preferable, in the author's opinion, is the numerical file. Reference to Fig. 4 will show that each anamnesis is numbered, the number on the history corresponding to a number on a card in a small card index system. The small card bears in addition the name of the