942 CHARLES E. DE M. SAJOUS 



quite inadequate grounds. There is no scientific advance without the use 

 of hypotheses and theories; indeed, they are among the greatest of the 

 stimuli that spur to accurate observation and experiment but only when 

 a sufficient aggregate of data warrants them. 



Pathologists and physiologists, though not averse to formulating the- 

 ories themselves, often upon insufficient grounds, sometimes infer that 

 the clinician has no right to construct hypotheses or to build up theories. 

 This, in my opinion, is a mistaken attitude. The worker in the clinical 

 sciences must avail himself of all the instruments of science just as does 

 the worker in the pre-clinical sciences. Thus only, in fact, will medical 

 progress by insured. 



Differences in capacity of different clinicians will doubtless be re- 

 sponsible for different methods of work. Some men can be most useful by 

 accumulating facts, others by arranging, comparing and trying to inter- 

 pret facts, still others by subjecting hypotheses to the crucial tests of fur- 

 ther observation and experiment. 



My own studies, first published in 1903 and 1907, were not true 

 hypotheses, but postulates reached after collating all available data (over 

 five thousand pro and con) contributed by the various branches of 

 medical science and personal experimental and clinical researches. All 

 these data were then analyzed and used for the upbuilding of the postu- 

 late reached, which thus became, so to say, the end-product of all knowl- 

 edge on the subject. 



Salient among the products of my labor, since confirmed by others, 

 were the following postulates: (1) that adrenal secretion is a factor in 

 the formation of oxyhemoglobin in the pulmonary alveoli and so is 

 significant for the processes of oxygenation and metabolism; (2) that 

 thyroid secretion is an important factor in the chemical processes of 

 immunization; (3) that trypsin is an active factor in cellular metab- 

 olism and immune chemistry; and (4) that the pituitary gland is neurally 

 connected with the suprarenal and thyroid glands and may influence the 

 secretory activity of these structures. Not only have these working- 

 hypotheses, based upon a very large aggregate of facts, biological, ex- 

 perimental and clinical, proved helpful to me in my own work, but they 

 have seemed to elucidate many physiological, pathological and clinical 

 data which previously had admittedly remained obscure. 



Each worker in endocrinology makes use of the hypotheses that 

 most appeal to him, but unless these be well sustained by collateral evi- 

 dence it tends to promote confusion. Before elaborating any theory an 

 investigator or clinician should consider all phases, pro and con of the 

 problem, and formulate his postulate as a result imposed by existing and 

 well grounded scientific data. Even should new facts compel him 

 to modify or abandon his hypothetical conclusion, the latter has served 

 its purpose if it has stimulated careful observation and experimentation. 



