458 Mr. G. A. K. Marshall on 



food in an entirely normal state, and it is notcwovtliy that 

 one out of the three was natahnsis. The extremely low 

 weight of the % natalcnHis in Expt. 10 also suggests some 

 unfavourable condition in the larval state. 



With the facts before us I do not see that any further 

 suggestions can be made at the present moment; but I 

 think the tables of weights clearly indicate the period at 

 which the stimuli should be applied, while temperature, 

 humidity, quantity or quality of food, or some combination 

 of these, seem to exhaust all probable influences in the 

 direction of a change of phase. 



I. Tke Bearing of the Seasonal Phases of Precis uiion the 

 Science of Insect Systematica. 



The results which have been described and illustrated 

 in this section of the present memoir are so startling that 

 they may well shake the confidence of naturalists in the 

 whole fabric of insect systematics. If such forms as 

 natalcnsis and scsainns, as sAmia and antilopc, as 'pcJasgis 

 and ari'hrsia, are nothing but the generations of two 

 alternating phases of a single species, approximately 

 synchronized with the heat and cold or humidity and 

 dryness of the alternating seasons, naturalists may feel 

 driven to ask, " What becomes of the validity of specific 

 distinctions ? " Between the two phases of Precis sesamus 

 there are extraordinary differences in colours, pattern, 

 shape of wings, relation of upper- to under-side, nay, even 

 in instinctive habits, including the choice of particular 

 stations. This latter distinction between the phases is 

 but the outward expression of some profound difference in 

 the intimate structure of ganglionic centres and inter- 

 com miuiicating strands in the central nervous system. 

 Important diiferences in venation are incidentally brought 

 about by the great differences in the shape of the wing. 

 The extreme rarity of intermediate varieties furthermore 

 recalls the abrupt transitions which are so common, 

 although very far from universal, between species of 

 animals which are assumed to be dirtinct. Under the 

 shock of Mr. Marsha ll's discovery that sesamiis and 

 ncdalensis are two forms of the same s^X'cies, the systematist 

 may well feel doubts about the foundations upon which 

 his science has been erected. In these distracting circum- 

 stances a firm belief in natural selection will be found to 



