360 FRAGMENTS OF SCIENCE. 



sess its facts and make them living seeds which shall 

 take root and grow in the mind, and not lie like dead 

 lumber in the store-house of memory. This is a task much 

 heavier than the . mere cataloguing of scientific achieve- 

 ments ; and it is one which, feeling my own want of time 

 and power to execute it aright, I might well hesitate to 

 accept. 



But let me sink excuses, and attack the work to the 

 best of my ability. First and foremost, then, I would ad- 

 vise you to get a knowledge of facts from actual observa- 

 tion. Facts looked at directly are vital ; when they pass 

 into words half the sap is taken out of them. You wish, 

 for example, to get a knowledge of magnetism ; well, pro- 

 vide yourself with a good book on the subject, if you can, 

 but do not be content with what the book tells you ; do 

 not be satisfied with its descriptive woodcuts ; see the 

 actual thing yourself. Half of our book-writers describe 

 experiments which they never made, and their descrip- 

 tions often lack both force and truth ; but no matter how 

 clever or conscientious they may be, their written words 

 cannot supply the place of actual observation. Every fact 

 has numerous radiations, which are shorn off by the man 

 who describes it. Go, then, to a philosophical instrument- 

 maker, and give, according to your means, for a straight 

 bar-magnet, say, half a crown, or, if you can afford it, five 

 shillings for a pair of them ; or get a smith to cut a length 

 of ten inches from a bar of steel an inch wide and half an 

 inch thick ; file its ends decently, harden it, and get some- 

 body like myself to magnetize it. Two bar-magnets are 

 better than one. Procure some darning-needles such as 

 these. Provide yourself also with a little unspun silk; 

 which will give you a suspending fibre void of torsion ; 

 make a little loop of paper or of wire, thus, and attach your 

 fibre to it. Do it neatly. In the loop place your darning- 

 needle, and bring the two ends or poles, as they are called, 



