398 MR. c. o'neill's experiments and 



a scale of 2 inches, divided into tenths, is laid down with a 

 diamond, a pair of forceps to lay hold of the hairs, and 

 two camel's-hair pencils moderately stiff; and the method 

 of working is as follows : — I lay out the sample of cotton 

 I am about to try, and take a pinch from half a dozen dif- 

 ferent parts, mix them together by pulling between the 

 fingers^ and then roll into a loose ball. I lay this cotton 

 upon a piece of black cloth, and the glass scale is also laid 

 upon black cloth ; then with the forceps I take up the first 

 fibre I put my eye on, and lay it on the glass plate. Every 

 ordinary fibre of cotton is thick at one end, and thin and 

 tapering at the other ; and as soon as I see which is the 

 thicker end, I lay hold of that part, but not quite at the 

 end, with the forceps in my right hand, and with a moist 

 camePs-hair pencil in my left hand, press gently upon the 

 fibre and with the forceps draw it between the pencil and 

 the glass until the extremity is over the first mark of the 

 scale ; I then roll the pencil over a little towards the thicker 

 end of the fibre, so as to take a good of hold it, drop the 

 forceps, and with the other camePs-hair pencil stroke the 

 fibre down on the scale, and read off its length to the 

 twentieth of an inch. I go all round the ball of cotton, 

 not choosing the hairs, but taking the first that offers to 

 the eye or hand : if the hair be evidently a broken one, and 

 has not the tapering extremity, I reject it ; but there are very 

 few of these in unmanufactured cotton ; it is usually the 

 other end which is broken. Cotton fibres are in a crumpled 

 state, and require some force to straighten them out, the 

 elasticity of good cotton being considerable ; but the glu- 

 tinous matter from the saliva with which the pencils are 

 wetted will hold them down on the glass pretty well : but 

 generally it is necessary to read off instantly the measure 

 of the fibre ; for the ends will curl up, if left, and the fibre 

 most likely be lost. I do not mean to say that I arrived at 

 this simple method all at once, or, even now, that it does 



