620 



♦ KNOWLEDGE ♦ 



[May 26, 1882. 



weight, V)ut it is fnared that they are too costly for tlio 

 average iimateur. Tliore are, however, several good makers 

 in Kiigland, such a-s Theiler, Blakcy, and others, whoso 

 magnets may 1)8 relied upon to support nine or ten times 

 their weight. Here it would not be out of place to mention 

 that tlie magnetisation of a piece of steel or iron depends, 

 not so much on the strength of the inducing magnet, as on 

 its own " saturation point " or its magnetic capacity. Tiie 

 purer the iron the higher is its saturation point, but this is 

 obtained at the expense of its power of retention, tiiat is to 

 say, pure iron cannot become a permanent magnet, although 

 it may be temporarily raised to a liigher state of magnetisa- 

 tion tlian a piece of steel or iron containing a very small 

 proportion of certain foreign substances (carbon, etc). 



Fig. 6 is a vertical section of the " pole-pieces " of our 

 TEagnets. B and C are two pieces of cast iron (softness 

 not essential, because of their being closely attached to, 

 and always actuated by, permanent magnets), each piece 

 being 4 inches long. They should be cast with a semi- 

 circular cavity, the two parts temporarily clamped and 

 carefully bored out to a diameter of about If or 1 J inches, 

 just large enough, in fact, to allow the armature to revolve 

 freely without touching. The edges (« e', dd), should be 

 filed down about an eighth of an inch, brass plates {p and p'), 

 having been previou.sIy fitted and tapped, ready for screw- 

 ing on to the pole-pieces, to render the combination com- 

 pact. G', G", G'", G"" are grooves, into which the magnets 

 are to be fitted, and should be just deep enough to make the 

 top and bottom of the pole-pieces flush with the magnets. 



Having secured the pole-pieces, the magnets should 

 be fixed. They should be of the U form, but not too 

 broad, a number of narrow ones put side by side giving 

 much better results. Referring to fig. 6, put all the 

 north poles in G' and G"", and all the south poles in 

 G" and G'". To fix them on, place small brass washers 

 top and bottom across the opening between each pair of 

 magnets, and pinch them together by small brass bolts and 

 nuts. We should recommend our readers to make their 

 own magnets if they possibly can, providing also that 

 they can procure sulficiently good steel. The steel should 

 bo in strips about a quarter of an inch thick, half to 

 three-quarters of an inch wide, and 9 inches long, bent so 

 as to measure 1 J to 2 inches between the legs of the U. 

 The best way to magnetise a strip, after it has been 

 brought to the desired shape, is to put coils of wire over 



the legs (in such a way as to form a continuous coil), 

 and pass a current of electricity through the wire 

 for a minute or so, breaking the circuit two or three times 

 during the operation. The magnets should be fitted as 

 close together as the bolts above referred to will permit, 

 but not touching, and should as nearly as possible fill up 

 the length of the pole-pieces. If any difliculty is expe- 

 rienced in this matter, Messrs. Blakey, Eminott, <fe Ca 

 (Halifax), will supply magnets ready to attach to the pole- 

 pieces at Is. Gd. each. Ten of these would be required, 

 that is, five on each side. There is a little more to be said, 

 which we must, however, defer till next week. 



THE COLD WEEK IN MAY. 



THERE is something startling in the placid way in 

 which M. de Fonvielle advanced, as we have seen, 

 the asteroidal theory of the cold week in iSIay is presented. 

 It was never held, even by Ertel, who first advanced it, to 

 be anything beyond a probable surmise, nor has it at any 

 time been adopted by astronomers of standing. But, in 

 reality, the supposed fact on which Ertel originally based 

 the theory, the only circumstance which gave to the theory 

 an appearance of plausibility, has been shown to be no 

 fact at all. It was formerly supposed that the stream of 

 small bodies, to whic i we owe the meteor shower com- 

 monly seen on Novem )er 13-14, is nearly circular in shape. 

 Now, one of the po nts where this stream crosses the 

 level of the earth's track corresponds with the place occu- 

 pied by the earth on November 13. Half a year 

 elapses between midnight, Nov. 13-14, and May 15, noon; 

 but, owing to the earth's more rapid motion in winter than 

 in summer, she crosses just opposite the place she had held 

 on Nov, 13-14 at midnight, on ^lay 12 about noon. Ertel 

 reasoned that, assuming the November meteors to form a 

 zone of small bodies around the sun, the zone being less iu 

 extent than the earth's orbit, this zone must of necessity 

 cross the level of the earth's orbit at a point directly 

 ojjposite the place of our November encounter with them, 

 and lying inside the earth's track, or between us and the 

 sun. The zone of meteors would therefore intercept a por- 

 tion of the sun's heat on or about May 11, 12, and 1 3. The 

 August meteors would, in a similar way, account for the 

 cold spell of February. And though the April " borrowing 

 days " could not be explained by a meteor system giving 

 October displays (for there is no such system), yet the 

 theory was not therefore invalidated. For, of course, a 

 meteor system may lie between our earth and the sun at 

 one part of the earth's annual course without necessarily 

 crossing that course itself at its opposite point. Thus it 

 seemed as though there were very strong positive evidence 

 in favour of Ertel's explanation of these cold snaps, and 

 no negative evidence of weight against it. 



This reasoning is often quoted at the present day, when 

 all its weight has departed from it. We now know very 

 certainly that neither the November nor the August meteor 

 systems pass between the earth's track and the sun on any 

 part of their circuit. The November meteors cross the 

 earth's track itself at the point she reaches on or about 

 November 13-14, and of necessity they cross the level of 

 her track again at a point e.xactly opposite, or lying in the 

 same direction from the sun as the earth does on or about 

 May 12. But, instead of this second place of crossing lying 

 between the earth and the sun, it lies far away in the 

 remote regions of the solar system, near the track of 

 the planet Uranus, or about twice as far from the sun as 

 Saturn, the remotest of all the planets known to ancient 

 astronomers. In fact, it is believed that we owe to the 



