146 FOREST PROTECTION 



crease of leaf-sulphur is not 

 due to an increase of soil- 

 sulphur (Haselhoff and Lin- 

 dau, p. 378); 



(2) to compare the analytic results 

 of such leaves and needles 

 only which were picked 

 equally old; 



equally situated with- 

 in the crown of the 

 trees; 



equally situated with 

 reference to eleva- 

 tion. 



All experts agree that short, sudden, strong 

 attacks by sulphur fumes are apt to be 

 deadly; still, such attacks do not cause 

 a VERY MARKED increase of SO 3 in the 

 leaves. 



On the other hand, long-continued, but slight 

 attacks by sulphur fumes result in a 

 heavy increase of SO 3 in the leaves; 

 still, such attacks do not cause a very 

 marked injury to the trees (Wislicenus, 

 Journal of Applied Chemistry, 1901, p. 

 28). 



It is evident, consequently, that conclusions 

 based on the chemical analysis of leaves 

 and needles are apt to be rash; and that 

 so-called chemical proofs must be viewed 

 with great precaution (Wieler, p. 380). 



D. UNRELIABILITY OF GLASS-CASE EXPERIMENTS. 



Experiments touching the poisonous effect of fumes made with 

 plants placed in a glass case cannot be so telling as experi- 

 ments made in the open, because: 



a. In the glass case, the gas is admitted from below so 



as to infest the lower surface of the leaves, which 

 lower surface is known to be more subject to 

 sulphur attacks than the upper surface. 



b. Sulphurous anhydrid, in statu nascendi, is increas- 



ingly active and pre-eminently corrosive. 



c. The discoloration of the leaves in nature differs from 



the discoloration usually observed in glass case 

 experiments. 



