THE JOURNAL OF PHARMACOLOGY. 105 



from it by one-tenth, or one-fifth of a degree. Any that show a greater difference 

 should be tested once more to make sure that no errror of manipulation has interfered. 

 If they still show a deviation of more than one-fifth of a degree they should be re- 

 jected. 



The same operation is then gone through for another point on the scale, and those 

 that finally pass this test are again arranged into groups, as before. Of course, those 

 that stand both tests equally well are the best and most valuable ones. 



Some thermometers have the peculiarity of apparently registering correctly, but 

 losing their register or index more or less rapidly when withdrawn from the source of 

 heat. For this reason it is necessary to re-examine them twenty-four hours (or, 

 if time is available, after a longer interval) after each of the before-mentioned tests, 

 when they should be found to offer the same reading as when they were first examined. 

 A retrocession of the mercurial index by as little as one- tenth of a degree renders the 

 rejection of the respective thermometer advisable. 



A further classification of the thermometers that have successfully passed all the 

 tests may now be made, to ascertain their sensitiveness. For this purpose they are 

 once more (but only once, at any optional part of the scale) tested as before described, 

 and removed from the warm water after an immersion of an exactly known period of 

 time. For instance, those that are found to have attained the correct temperature 

 (according to the standard which, if sluggish, may have been previously inserted for 

 a sufficient length of time), after the lapse of sixty seconds or one minute, may be set 

 aside as one-minute thermometers. Similarly, there may be separate groups of 2-, 3- 

 (or more) minute thermometers, which will often be found useful, as many purchasers 

 lay value on the time within which a thermometer responds. 



In making contracts for clinical thermometers there are many features which per- 

 mit an exact definition or description, so that a control of the articles delivered is not 

 difficult. The only feature that is not definable or controllable is the age or ripeness 

 of the thermometers. Nothing beyond the word of the maker can tell whether they 

 had been allowed to become seasoned or not. It may be of interest to some of the 

 readers of this paper to know what conditions are prescribed for the clinical thermom- 

 eters bought, under contract, for the public hospitals of the City of New York. The 

 specifications read as follows: "... gross of Clinical Thermometers (to he de- 

 livered in installments as required), 4-inches long, to be substantially made, with single 

 bulb, plain front, indestructible index, each even degree plainly numbered, the gradu- 

 ation between 94° and rio° F. extending over a space of not less than 1% inches, and 

 to be correct within 0.2 of a degree as determined by the standard thermometer of the 

 Departmant." It was found useless to add a condition as to sensitiveness, because 

 among a gross or more delivered at one time there are always found a number which 

 are more sensitive than others, and these are reserved for special purposes. The 

 higher sensitiveness is mainly due to the fact that the glass of the bulb is thinner, for 

 which reason these sensitive thermometers are much more easily broken. For all 

 ordinary purposes a good solid 3 or 4-minute thermometer is much preferable. 



Thermometers remaining in stock should be examined again at the expiration of 

 every six months, and a record should be kept of the number and condition of all those 

 which have retained their accuracy after the first six months. After every further 

 lapse of six months, if they are still found to be accurate, their value will have pro- 

 portionately increased, as they may then be certified in good faith as being seasoned. 



New York, September 13, 1898. 



Book Reviews. 



Catalago de Plantas Mexicanas Fanerogamos. Arreglado par el Dr. Manuei, 



Urbina, Imprenta del Museo Nacional, Mexico. 



This is a large octavo of 487 pages, listing some 3.000 plants, arranged according to 

 the Benthamian system. The botanical names are followed by citations of their pub- 

 lication, important synonyms, the habitat, locality where collected, with the names 

 of the collectors and their collection numbers, when any. In many cases common 

 names are appended. No systematic attempt appears to have been made to fix upon 

 the accurate name, various leading authorities having been followed, with results not 

 altogether consistent. The publication, however incomplete, must find an important 

 use. H. H. R. 



