194 



NATURE 



[February 27, 1896 



thoroughly well undertaken, must not be restricted simply 

 to the species of tree, but should include the conditions 

 ,of their position and surroundings, such as soil, moisture, 

 &c., as these may play an important part. 



The only operations, as far as we know, that seem to 

 have been systematically carried out on these lines, are 

 those which since the year 1874 have been made by the 

 furstlich Lippe-Detmoldschen Forstdirektion. 



The statistics show that from 1879 to 1890 lightning 

 liad struck 56 oaks, 3 or 4 pines, 20 or 21 firs, but not a 

 single instance of a beech tree was recorded. These facts 

 will be seen to be of importance when it is stated that 

 the relations between the numbers of different* kinds of 

 trees in the region under observation were such that, out 

 of 100 trees, about 70 were beech, 11 oaks, 13 pines, and 

 6 were firs. The numbers show at a glance that beech 

 trees seemed to have been entirely free from attack, 

 although they were twice as numerous as all the other 

 trees put together. . A practical hint can at once be 

 deduced from this ; for protection against lightning, 

 when one is in, perhaps, a wood, can be apparently 

 secured, provided of course there are beech trees 

 there! 



In the investigation of the underlying causes, it was 

 .considered unnecessary to take into account the nature 

 of the soil, for observation showed that trees situated on 

 wet ground were very often struck, thus contradicting 

 the generally conceded idea that this only occurred when 

 they were on soil which contained a very small amount 

 of water. Further, it was also diilficult to trace any con- 

 nection between the different depths to which the roots 

 of trees extended, since this is a subject about which very 

 little is known, and no statistics are at hand for trees that 

 have been thus damaged. 



The next step was to consider the species of tree, and 

 investigate whether any facts here might be brought to 

 light. With this point well in mind it was necessary, first, 

 to determine the degree of conductivity of the different 

 kinds of wood. Experiments in this direction have 

 previously been made by Du Moncel, but his observ- 

 ations, for many reasons, gave very variable values, so 

 variable in fact that he himself could not lay any weight 

 on them. 



One fact known, with regard to the conductivity of 

 different species of wood by the help of the electric spark, 

 is that electricity is conducted better in the direction of 

 the grain than in the direction perpendicular to it. 

 Further, Caspary has shown, in the case of lime- and dry 

 pine-wood, that conductivity varies according as the 

 direction is longitudinal, radial, or tangential. 



Notwithstanding the results just mentioned, Jonescu ^ 

 undertook the work afresh, and made independent 

 observations, employing the electric spark ^throughout, 

 and taking into consideration the anatomical, chemical, 

 and physiological properties of the different kinds of 

 wood. 



It will be sufficient for us to deal simply with the results 

 of these investigations, without entering into the method 

 of experiment adopted. 



The first experiments with oak and beech indicated 

 that the former was a very much better conductor than 

 the latter, just as appeared to be the case from the statis- 

 tics given above, if conductivity be assumed to play the 

 most prominent part. Similar experiments were made 

 with black-poplar and willow. With living wood— that 

 is, wood just cut, and therefore containing moisture — the 

 percentage of the latter being, according to Schiibler and 

 R. Hartig, for black-poplar 5 r8, beech 397, oak 35-4, 

 and willow 26*0, experiments led to the result that con- 

 ductivity of wood in this state is independent of the 

 amount of moisture inside. 



1 Berichten der Deutschen Botanischen Gesellschaft. Bd. xii., 1894, 

 S. 129-136. 



NO. 1374, VOL. 53] 



Extending the observations over a far wider range, and 

 employing numerous different kinds of woods containing 

 varying quantities of fatty materials, such as oil and 

 resin, it was discovered that the wood cut from living 

 trees was in every case a worse conductor of electricity 

 the more oil or resin they contained. The fresh wood of 

 trees, on the other hand, which are rich in starch but poor 

 with regard to fatty matter, conducted electricity very 

 well, although no large difference for the various kinds of 

 wood was noticed. 



It will thus be seen that the question under considera- 

 tion seems to reduce itself to the finding out the qualities 

 of the juicy matters in the wood of the trees, as these 

 seem to play an important, if not the whole, role in the 

 proceedings. 



As regards the distribution of the fatty materials and 

 the starch in the wood of trees, we have to apply to Fisher 

 and Suroz, who, from their investigations on these points, 

 have shown that the quantity of oil and starch varies 

 with the time of year ; th'ey have,, further, classified trees 

 according to their richness in these materials. 



According to these facts then, assuming in the case of 

 Germany jjiat it is during the summer months that 

 thunderstorms are most frequent, those trees rich in fatty 

 materials (Fettbaume), and which during the summer 

 contain much of them, are to a great extent protected 

 against lightning. Those, on the other hand, that are 

 poor in oil during the period of thunder-storms, and 

 especially such trees which contain much starch (Stark- 

 baume), are more liable to be struck. 



One very interesting example to which reference is made 

 is the wood-pine. This tree, during the summer months in 

 Germany, is, comparatively speaking, very often damaged, 

 while in countries where the thunderstorms occur in the 

 winter months (Ireland and Norway) it is usually un- 

 touched. To explain this according to the deductions of 

 Jonescu's observations, it must be shown that in winter 

 the tree is richer in oil than in summer. A microscopic 

 examination of samples at both these periods of the year 

 has proved conclusively that this is really the case, the 

 oil disappearing in April, to have its place taken by 

 glucose and starch. Employing the apparatus previously 

 referred to, Jonescu found that for the winter wood of this 

 tree double the amount of potential energy was required 

 for sparking purposes than was required when operating 

 with the summer wood. 



As a check on his own results, Jonescu took the wood 

 of typical trees rich in fatty materials (Fettbaume), beech 

 and walnut, and found that when deprived of their oil by 

 means of ether, their degrees of conductivity were in- 

 creased and became practically the same as those of the 

 wood of typical trees rich in starch (Starkebaume). 



The above explanation of the causes which render 

 some trees more than others liable to the destructive 

 power of a lightning flash, helps certainly to explain 

 other points of less importance. Experiment showed 

 that for both kinds of trees, namely, those rich in fatty 

 materials and starch respectively, the wood in the living 

 state is a far worse conductor than when dead. This 

 leads to the deduction that trees with dead branches on 

 them are more likely to be struck than those without any, 

 so that they should be avoided if possible. Jonescu's 

 hypothesis also explains why it is that lightning does not 

 as a rule strike the highest part of a tree, but generally 

 the trunk, either inside or underneath the crown. 



In conclusion, it is to be hoped that these investigations 

 of Jonescu, to which only brief reference here has been 

 made, will be followed by observations and statistics from 

 other sources, fqr this question is one that is of very 

 general interest. It is sad to think, however, that the 

 oak, a tree which we cherish so much, is such a friend to 

 flashes of lightning in the way of conduction, that it in 

 this way brings on its own destruction. D. 



