4798 Fishes. 



On the 22nd July, 1833, six parr were caught with the artificial fly 

 at Romaro bridge on the Lyne, a branch of the Tweed. They were of 

 the usual size, averaging probably 5 inches : all were males, with the 

 milts large, and 2j inches in length. Do we usually find salmon with 

 the milts large in July ? And if these parr become smolts next May, 

 what in the meantime becomes of the enormously enlarged milts? 



On the 3rd September, of thirteen parr caught in the Tweed, 

 between the Bield and Polmudie, there were only two females, and 

 such observations, if repeated and supported by extensive evidence, 

 might lead to important results. In the mean time, it is remarked in 

 my journal, that of these eleven males some were 8J inches in length, 

 others only 4 or 5 inches ; in the larger the milts were enormously 

 enlarged. Now, what became of these males, which in September 

 were as large as the largest May smolts ? Did they cease to grow 

 until the arrival of May ? What became of the milts ? And if 

 such a premature development of the male takes place without a 

 cause and without an object, how comes it that no May smolt ever 

 shows traces of these organs having been once developed ? 



In conclusion, it is to be regretted that experiments which might 

 have formed valuable acquisitions to science have been so conducted 

 as to be repudiated by most scientific men. Profoundly ignorant of 

 the basis of all zoological science, the observers could not be trusted. 

 Those being now conducted near Perth are in quite different hands, 

 and will, no doubt, lead to important results. The question of a dis- 

 proportion of males to females, as regards the autumn and winter parr, 

 did not escape my notice, as may be seen by reference to page 93 of 

 the work so often referred to in this brief notice ; at page 95 will be 

 found a minute account of the experiments on the Dee, proving the 

 three-year old theory. With a little more confinement and restraint, 

 the period of development of the smolt might perhaps have been ex- 

 tended to four years ;* the two and three-year theories have now 

 been disposed of by my friend Dr. Esdaile, but the one year view has 

 still to be proved, and this can never be satisfactorily done by breed- 

 ing in artificial waters. 



R. Knox. 

 M lessen House, Upper Clapton, 

 June, 1855. 



* Mr. Newman, editor of the ' Zoologist,' and a distinguished naturalist, informs 

 me that you may keep tadpoles as tadpoles as long as you like, merely by restraining 

 their development. 



