40 



taiice in their economy. The large diverticuhnn from the stom- 

 ach is, no doubt, to this end. In certain Homopterous families 

 a filter is formed between the aesophagus and the rectnm 

 through which the excessive juices pass, only the more nutritive 

 substances passing through the whole alimentary canal. In the 

 case of the "Frog-hoppers" the young stages are passed in some 

 of the secreted liquids, and it has been suggested that this forms 

 a protection to them, and has been brought about by I^atural 

 Selection. It has also been suggested that the liquid surround- 

 ings are necessary for the protection of the tender bodies against 

 atmospheric changes. In reply to the first suggestion it must be 

 remembered that the young insects are heavily attacked by their 

 enemies, to whom the white mass of liquid only serves as a 

 guide. In regard to the second suggestion there is good reason 

 to believe that these insects descended from ancestors M'hose 

 nymphs did not live in "spittle" and had normally hardened 

 integuments. To me it appears more feasible that the secretion 

 is due to the feeding habits, and the soft integument a direct 

 result of those habits. 



Isolation. 



The Rev. John T. Gulick in his writings* on this subject, 

 on which Romanes placed "a higher value than any other worl; 

 in the field of Darwinian thought since the date of Darwin's 

 death", has shown that geographical and topographical isolation 

 has played the chief part in the evolution of Hawaiian land 

 shells. These writings should be better known to Hawaiian 

 entomologists than they are, as I believe still further evidence 

 can be found for these theories among the insects. In this con- 

 nection it is of interest to consider the numerous species in some 

 of the flightless genera, a condition that facilitates isolation. 



Romanes and Gulick have shown that Natural Selection 

 without the aid of some other form of isolatiou leads only to 

 monotypic evolution, but with the help of physiological isola- 

 tion Romanes considers that it can lead to polytypic evolution. 

 This is a point that I cannot agree with. If a variety arises 

 with a character which gives it a better chance of existence than 

 the parent species, and if the character follows the Mendelian 

 law of inheritance and so does not become "swamped", it will 



* Journal Linn. Soc. (Zoology) XX and XXII. 



