448 TRANSACTIONS OF THE AMERICAN PHILOSOPHICAL SOCIETY 



may be the chief cause of the striking resemblance of the natural mechanism of the fish 

 skull to machines designed by human minds. I have already used similar considerations 

 in support of the conclusion that human minds differ from fish minds only in complexity 

 and that both are dependent upon the anticipatory values of the nervous system. Doubt- 

 less a certain measure of hereditary variability in likes and dislikes has been prerequisite 

 for progressive changes in any given line of fish skulls. 



"Basic Patents" and Evolution 

 We have seen that in many lines of descent the excellent "basic patents" of the per- 

 comorph skull have often served as the starting-point for new and wholly unexpected 

 devices, such as the nutcracker jaws of scarids or the folding scoop-net of Stylophorus. 

 On the other hand, a new and successful invention often begins to lose its distinctive 

 characteristics and may finally degenerate and disappear or become disguised so that we 

 can only recognize its origin after prolonged study. Thus the protractile apparatus of 

 typical percoids loses its protractility in many scombroids and other groups, the long jaws 

 of the primitive scombresocid become greatly shortened in the adult fiying-fishes, the 

 exquisitely designed illicium of the central pediculate degenerates in the sea-bats, and so 

 forth. 



Fish Skulls and Natural Selection 



The foregoing studies afford some evidence that, considering the bony fishes as a whole, 

 the mouth, jaws and teeth have varied in all directions according to the curve of probability, 

 limited by certain fixed necessities of a flexible gateway to the digestive system. Of all 

 the conceivable modifications of the jaws and teeth, a surprisingly large number have proved 

 to be useful at different times during the hundreds of millions of years of the geologic record. 

 The student of deep-sea fishes also will readily agree that almost every conceivable com- 

 bination and apparent misfit of grossly disproportionate development of certain parts has 

 occurred, perhaps under the distorting stresses of great cold and darkness, and that a sur- 

 prisingly large number have proved either useful or not fatal to their possessors. But it 

 should again be noted that within the hosts of the teleosts apparently random changes 

 have never gone so far as to destroy such "basic patents" as the suspension of the jaws by 

 the hyomandibular, or the functional integrity of the quadrate-mandibular joint, or the 

 articulation of the hyomandibular with the pterotic, etc. Nor is any record known to me 

 of an individual fish that has been congenitally deficient in these features. Such basic 

 patents have thus been kept intact during the entire history of the ganoids and teleosts 

 from the Lower Devonian to the present day, — a period of nearly four hundred million 

 years. 



It is significant that the researches of the physicists upon the ages of various rocks have 

 multiplied by twenty-fold the earlier estimates of the length of time in which it was formerly 

 thought that the observed evolution of fishes had taken place. In other words, Nature, 

 operating on small or large new hereditary tendencies, has had perhaps a hundred million 

 years for her experiments, in order to change moderately sized jaws either into nibbling 

 jaws or into "scoop-nets"; while for the entire transformation from the agnathous to the 

 percoid type there may have been available as much as four hundred million years. 



Such vast time periods imply equally enormous numbers of variably qualified individ- 

 uals, conserving by heredity the advantageous characters of the past, subject in the long 



