TYPHO-ID FEVER. 387 upon the stage of a microscope and examined for the agglutinations. As recent extended observations have shown that occa- sionally the blood of healthy men and animals has the power of producing the agglutinations, the consensus of opinion now seems to be in favor of the view that a cer- tain dilution of the blood is required for a satisfactory diagnosis, and that all reactions with concentrations greater than one part of blood in fifty of culture may be questionable, while less concentrated dilutions are almost positively diagnostic. A time-limit must be placed upon the experiment. For the weak dilution not mpre than two hours should be required for a perfect reaction, and for the stronger solution correspondingly less time should be required. A curious fact that should not be overlooked is that the agglutinating substance is not constantly present in the blood, but sometimes alternates, being present for several days and then absent for a day or two. The agglutinating power of the blood occurs early in the course of typhoid, and in typical cases seems to be present in the first week of actual illness. A point that should not be forgotten is that the agglu- tination of the bacilli seems to be a phenomenon quite in- dependent of any immunity possessed by the individual, and therefore is not an "immunity-reaction." Just what the agglutinating substance is, has not yet been determined. The agglutinations are occasionally caused by the serum and dried blood from other diseases than typhoid, but in a collection of 4000 cases it was shown that the errors from this source were only about 5 per cent. Malvoz l has experimented with a number of chemicals, and has found that formaldehyd, corrosive sublimate, per- oxid of hydrogen, strong alcohol, and anilin colors (such as chrysoidin, vesuvin, and safranin) have the power to produce the typical agglutinations even in very dilute solutions. 1 Ann. de T Inst. Pastettr, xl., 7, 1897.