60 J. E. Marr and E. H. Adie — The Lakes of Snou-don. 



parallel with those of the contour-lines above, where they run round 

 a little valley occupied by a small stream. The existence of this 

 bay is, of course, explicable if the lake be drift-dammed, but 

 is difficult to explain if we suppose that it has been excavated 

 by ice. 



The colour of the water of Glaslyn is indigo, though the tint is 

 not so deep as that of Llyu du'r Arddu. Here also copper-mines 

 have been worked close to the lake, but, as has been mentioned above, 

 no trace of copper was found in the water. 



The samples, of which the analyses are appended in tabular form, 

 were obtained under somewhat different conditions in the case of 

 each lake. That from Llyu du'r Arddu was obtained from fairly 

 deep water, surrounded on the landward side by rock in situ, 

 that of Llydaw from the middle of the causeway which has been 

 made across the lake, whilst that of Glaslyn was obtained from 

 shallow water close to an ordinary foreshore, consisting of loose 

 blocks and some vegetation. 



ARDDU. LLYDAW. GLASLYN. 



Farts per million. 



Solids }I°o^g^'ii« 



( Organic 



Hardness (Lime and magnesia salts) 



Chlorine 



Nitrogen as ammonia 



,, albuminoid ammonia ... 



Oxygen absorbed in 15 min. 



„ ,, 4 hrs. 



Nitrogen as nitrates and nitrites ... 



We did not think it necessary to determine the organic carbon and 

 nitrogen, as the above results may be generally used as a substitute 

 for them. 



The analyses show that the waters of Arddu and Llydaw are 

 similar in character : they contain about the same amount and variety 

 of inorganic matter (chlorine, lime, etc.), but Llydaw contains rather 

 more than twice the amount of organic matter that Arddu does, as 

 shown by the albuminoid ammonia, and oxygen absorption in four 

 hours. The similarity of the amount of chlorine also suggests that 

 the organic matter is similar in the two cases. This result is what 

 would be expected from the relative position of the lakes. 



The results in the case of Glaslyn are very remarkable. The 

 amount of solids (not lime and magnesia salts) and chlorine, also of 

 ammonia (free and albuminoid), oxygen absorbed, and nitrogen as 

 nitrates, would always be held to indicate the presence of much 

 animal organic matter. This hypothesis seems at first ridiculous 

 from the position of the lake, unless the hotel on Snowdon summit 

 drains in any way into it. The only other interpretation is, that the 

 sample obtained from near the bank was not of average quality. 

 No impurity of such character could be introduced in any other way. 



The object of these analyses, viz., to put to the proof M. Forel's 

 hypothesis of the cause of coloration of mountain lakes, is unfor- 

 tunately not attainable from these results, owing to the peculiarity 



