52 



SCIENCE. 



[Vol. V., No. 102. 



impossible to directly restrict or control these scien- 

 tific operations by law. The general purpose of the 

 work may be formulated in the statutes, and the 

 operations may be limited by the appropriations 

 made therefor, and this is as far as the statute itself 

 can properly go; for, if the operations themselves 

 could be formulated in law, the facts would already 

 be known, and the investigation would be unneces- 

 sary. It being impossible by statute to control or 

 restrict the lines of investigation, as above shown, 

 there is yet a control of the official personal organiza- 

 tion which can properly be exercised by statutory 

 provision ; and a further control, superior to the im- 

 mediate organization prosecuting the work, may be 

 properly exercised in relation to the financial opera- 

 tions in the payment of employees, and in the pur- 

 chase, use, and custody of public property, and the 

 supervision of accounts. 



I beg permission to set forth certain facts, which, I 

 think, should be used as a guide in the establishment 

 of such official organization ' and superior control. 

 In the first place, the investigations prosecuted by all 

 of these scientific institutions are in their nature in- 

 ter-related and interdependent. The success of one is 

 dependent, to a large extent, upon the success of the 

 others ; and, if at any time in the correlated investi- 

 gations prosecuted by the general government any 

 one branch fails in its department, the other branches 

 suffer therefrom to a greater or less extent. 



Forexample : geodetic operations carried on through- 

 out the world, and having for their purpose the de- 

 termination of the figure of the earth, were for a long 

 time prosecuted by refined trigonometric methods; 

 but, as the work progressed, the problem was found 

 to be more complex than was at first supposed, and 

 elaborate gravity determinations were added to trigo- 

 nometric methods. And it has quite lately been dis- 

 covered that trigonometric and gravity methods must 

 yet be supplemented by the determination of the 

 geologic structure of lands, especially of mountains 

 and mountain systems. Thus it has been found that 

 the geographer cannot accomplish his work without 

 appealing to the geologist for his knowledge. On the 

 other hand, it has been found in the study of struc- 

 tural geology — and by that is meant the plan upon 

 which the rocks composing the lands of a country are 

 arranged — that it cannot be clearly understood and 

 explained without the facts of geodesy. Sound geo- 

 logic research, therefore, must progress hand in hand 

 with sound geodetic research. 



Again: in the prosecution of geodetic research, the 

 parties thus engaged determine the exact position in 

 latitude, longitude, and altitude, of many points upon 

 the surface of the earth. In the prosecution of a geo- 

 logic survey of the same territory, these same points 

 must also be known ; but, more than that, their num- 

 ber must be vastly multiplied, so that a map may be 

 constructed setting forth the latitude, longitude, and 

 altitude of all portions of the country surveyed. 

 Where the geodetic survey establishes but hundreds 

 of points, the geologic survey must have millions of 

 points established. 



Again : the points to be used in the geodetic survey 



must necessarily be selected for that purpose. A gen- 

 eral reconnoissance of the country over which such a 

 survey is carried must be made, and the materials 

 collected for at least a skeleton map. Thus it is that 

 a skeleton map is necessary for a geodetic survey, and 

 a completed map for the geologic survey. In like 

 manner it can be shown that the relations between 

 geodetic and geologic work are manifold, and, still 

 further, that the geodetic work and the geologic 

 work have a great variety of connections with the 

 other scientific works prosecuted by the general gov- 

 ernment. It would require a volume to set forth all 

 these relations, and to show how completely the suc- 

 cess of one is dependent upon the success of all. 



It will thus be seen that the official organizations 

 for these institutions should be co-ordinated, that 

 they may work together and aid each other; and, 

 further, as each is interested to a greater or less ex- 

 tent in the operations of the other, the organization 

 should be such that one shall not be compelled to do 

 that which is the proper function of another, and 

 that no one shall be permitted to encroach upon the 

 functions of another. As long as the several scien- 

 tific commissions and bureaus of the general govern- 

 ment are distributed through all the departments of 

 the government, — one in the war department, anoth- 

 er in the navy, another in the interior, another in the 

 treasury, etc., — each bureau must necessarily, to a 

 large extent, be autonomous: they must be self-gov- 

 erned, for it is a practical impossibility for any secre- 

 tary of a general department to make such a study of 

 the methods of scientific research as would warrant 

 him in attempting their control. Hence these insti- 

 tutions have in the past been to a great degree auton- 

 omous, and must, under the same plan, continue to 

 be. 



If the statements thus briefly made are correct, it 

 follows that the first guiding principle to the proper 

 official organization of the scientific work is as fol- 

 lows: The scientific institutions of the government 

 should be placed under one general management. 



Again : as a necessity, scientific investigation must 

 be controlled by the facts discovered from year to 

 year, and from month to month, and from day to 

 day. The operations of investigation, therefore, can 

 only be controlled by the men who are actually per- 

 forming the work. For example : the director of the 

 geological survey cannot possibly lay out the work 

 for his assistants in detail. He can only set forth in 

 a general way the object to be reached, the general 

 methods to be pursued; and such plans must be held 

 open to revision from time to time as the facts discov- 

 ered by the investigators themselves may demand. 

 He must therefore hold himself always in commu- 

 nication with his assistants, and ever be ready to 

 entertain their suggestions; and there is always a 

 probability that he will err more in the direction of 

 rejecting wise suggestions than accepting unwise 

 plans. 



It is thus that, to a large extent, the plans of the 

 work prosecuted by an organization for scientific re- 

 search must originate with the experts and specialists 

 who are themselves engaged in the investigation ; and 



