METHODS OF EESEARCH! PREPARATION OF ILLUSTRATIONS 123 



natural size with the ordinary lenses and an extra long bellows, 

 or by means of planar enlargements. These latter enable one 

 to show distinctly even the minutest color changes, or form 

 changes, on leaves, stems, fruits, or other organs. 



Frequently when a very good photograph has been furnished 

 the engraver, he ruins the plate he has made from it by over- 

 etching. In such cases he should be compelled to make another. 

 But the best half-tone ever made can be ruined by a second rate 

 pressman. For an example of a good plate ruined by a slovenly 

 pressman see ''Bacteria in Relation to Plant Diseases," Vol. I, 

 page 178, and for a properly printed half-tone from the same 

 negative, see Fig. 67 in this book. 



Enlargement to twice natural size requires the ground glass 

 (focusing plate) to be from the lens a distance equal to three 

 times the equivalent focus of the lens, i.e., the enlargement of an 

 object to twice natural size, using a lens of 9^^ inch focal length, 

 requires the placing of the object 14)^ inches in front of the 

 lens and the bellows extended to about 28 inches. 



Photomicrographs. — ^Success with the large horizontal photo- 

 micrographic apparatus formerly much used by the writer 

 depends on keeping a few things constantly in mind: (1) the 

 source of light must be reasonably constant ; (2) the rays of light 

 must pass in a straight line from the luminary to the ground glass, 

 i.e., every piece of the apparatus must be centered; (3) a ray filter 

 should be used; (4) if the bellows is elongated to twice its former 

 length the exposure must be quadrupled, and vice versa; (5) 

 if high power lenses are substituted for low power ones the ex- 

 posure must be lengthened (one comes after a time to judge 

 very correctly of the exposure by the amount of light on the 

 ground glass, but the beginner will save much time and vexation 

 by using an exposure shutter, and in its absence by drawing 

 the ordinary slide ^i, then i^, ^i, and finally entirely, and judg- 

 ing by development with a normal developer which of the various 

 exposures, or between or beyond which, will give the best re- 

 sults) ; (6) the time required for unstained and thin-stained slides 

 is very much less than that for those heavily stained, and in 

 exposing thick or dark-stained slides the time must be long enough 



