A few years ago I bought an old medical dictionary (The American Illustrated Medical Dictionary, 20th edition, printed in 1946). I soon discovered that it wasn't very helpful because terminology has changed so much. Looking things up on the internet was usually both easier and better. Occasionally I'd use it to look something up if I didn't have the internet available or I'd flip it open & read a few entries when I was bored.
It has 1668 pages and boasts 885 illustrations. Of the 885 illustrations some are portraits of notable people (which is a pretty much a waste of space in a dictionary, I'd say). A random page has 44 entries on it. Assuming that is typical then there are about 73,000 entries in the book. The title page notes that 240 illustrations are portraits leaving 645 useful diagrams. That is 1 useful illustration for every 113 entries. One would imagine that scarcity would drive the editors to come up with criteria that would insure only the most significant entries and the entries that could best be explained with an illustration would get an illustration. On the other hand, someone might insist that 4 different types of hymens absolutely must be illustrated. Oh, and that the entry on elephantitis should definitely include an illustration of elephantitis of the scrotum.