December 2, 1921] 



SCIENCE 



549 



musemn. Undoubtedly many come merely 

 to toy with the apparatus, but some few pore 

 over the explanations and ask questions about 

 them. That it has awakened an interest in 

 the subject in many for the first time may 

 be taken for granted. One very definite ad- 

 vantage is that it allows the instructor to 

 refer his students to certain experiments in 

 the museum with the request that they try 

 them and report on the results, e.g., all our 

 elementary students determine, from its 

 period, the length of the large pendulum. 



However, while it seems eminently worth 

 while it is needless to say that such a museum, 

 simple as it is, will not run itself. Although 

 it does not require the presence of an attend- 

 ant, its continued demand for new experi- 

 ments as well as the upkeep of the old ones 

 would constitute a perhaps unwarranted li- 

 ability on the time of the instructional force 

 of the department if it could not, as in the 

 present case, be entirely turned over to an 

 ingenious and able apparatus man. 



L. E. Ingeesoll 



Madison, Wis., 

 November 5, 1921 



HOW TO DO RESEARCH i 



I HAVE never done any research. I am 

 therefore able to give unbiased advice re- 

 garding it. 



Besearch — in the broadest sense — consists 

 largely of repairing leaks in glass tubing. 



More specifically, it consists of gathering 

 in a cell down in the Eyerson basement a 

 weird assembly of switches, wires and glass 

 tubing — and then keeping other students 

 from borrowing it. 



Apparatus may be borrowed or acquired. 

 If you borrow it you are expected to return 

 it. If you acquire it, you keep it until you 

 are found out. 



Tools at one time could be found in the 

 student's shop. Now you find them every- 

 where. 



1 Read at a gathering of the graduate students 

 in Physics of the Eyerson Physical Laboratory on 

 a social occasion preceding Professor Millikeu 's de- 

 parture from Chicago. 



In order to do research, one must have 

 ideas. One idea is sufficient. Two ideas are 

 apt to contradict each other. 



Ideas are easy to get. If you haven't any, 

 consult Dr. Gale. He can be found adjust- 

 ing gratings down in the basement. 



By all means do not search for something 

 original. If you think you have a new idea 

 read Professor Groszkopf 's articles in " Zeit- 

 schrift fiir So und So " published about 1700. 

 You will find he suggested the same thing 

 two centuries ago. 



After all, it is doubtful whether even one 

 idea is necessary. Merely get some appara- 

 tus, solder it together and take readings. 



Eeadings are always taken through a tele- 

 scope. 



You will get certain numbers. Plot these 

 numbers against other numbers which you get 

 from variable parts of the apparatus. 



If you get a straight line on plotting your 

 observations you know at once that the re- 

 sults could have been predicted. 



However, if you get a curve the situation 

 is different. Examine the curve carefully for 

 sharp bends or breaks. If you find one, you 

 have made a discovery. These breaks are 

 significant. Consider carefully what may 

 have caused such breaks. Try to trace them 

 to atomic or electronic phenomena. Draw a 

 picture of the atom. Don't be discouraged 

 if your picture doesn't agree with other pic- 

 tures. Dr. Lunn will show it doesn't mean 

 anything anyhow. 



Having obtained a curve and concocted a 

 theory, it is befitting that you present the 

 whole to the Physics Club. 



The Physics Club was invented to keep 

 research students from getting the big head. 

 It consists of a crowd of professional knock- 

 ers. There is one booster. You are the 

 booster. 



It is fitting here to give you details on 

 your conduct at the meeting. 



The latter is always preceded by tea. While 

 this is being served go into the lecture room 

 and copy a few weird sketches of your ap- 

 paratus on the board. Make everything as 



