LOOKING UPWARD IN DUSK AND DARK 43 



caught at a disadvantage with its wings half out, or, again, 

 it might not, but I cannot think of anything else. 



I was not below in the observation chamber to observe 

 the effect of a floating Coch-y-bondhu made in the old style, 

 or a Kennedy's floating Coch-y-bondhu beetle. Our last 

 experiment with the trout fly was with a good-sized floating 

 Sedge, but I recall nothing new or of interest about it. 



I recognize that it is not safe to dogmatize or deduce 

 very much from these very incomplete, very brief, and 

 very imperfect observations, made by one like myself 

 not equipped with the scientific knowledge to draw the 

 inevitably right deductions from them, and the little which I 

 have since observed does not take me much further. I do not 

 know the true meaning of the structure of a trout's eye. I 

 cannot tell what may have been the disturbing or distorting 

 effect of the sheet of plate-glass between me and the water. 

 And there are doubtless many other factors I have not 

 allowed for, and I recognize that the course of experiment 

 ought to be pursued systematically for weeks and months 

 and years in all sorts of lights and all sorts of weather 

 before any safe deductions can be drawn. So my readers 

 (if any have got so far) will understand me that I am putting 

 forward this record of my observations, not as establishing 

 anything, but as containing perhaps some suggestions for 

 investigation which others, more fortunately situated and 

 better equipped than I am, may be able to follow up and 

 verify or disprove. 



XII 



LOOKING UPWARD IN DUSK AND DARK 



Under another heading we have considered the trout's 

 vision looking upward in daylight. Let us now try and 

 put ourselves in his place after the sun has gone down 



