416 SECTIONAL TRANSACTIONS.—G. 
further that only direct lightning strokes closing on the line conductors, 
tower or ground wires need be considered, induced surges only assuming 
importance on low voltage lines. 
The dependence of the occurrence of a flashover across an insulator 
string from a stricken tower or ground wire to the line conductors on the 
stroke current and the tower foot resistance is then discussed, and from the 
equations included it is shown how flashovers may be prevented by the use 
of low values of the earth resistances or by the use of counterpoises. As yet 
the actual value of the rate of rise of lightning voltage at a stricken point is 
not known, but it is reasoned that particularly on medium voltage lines 
many flashovers may be initiated by breakdown across an insulator string 
before the return of the reflected wave from the tower foot by reason of a 
rapid rate of increase of lightning voltage of the order of several thousand 
kilovolts per micro-second. 
The influence of direct strokes and flashovers on line behaviour both 
from the point of view of power frequency follow up current and the 
propagation of travelling waves is then considered. ‘The effects of the 
latter, together with those of direct strokes to shielded and unshielded sub- 
stations, is discussed in relation to internal stresses in transformer windings, 
and it is pointed out that the most severe are axial stresses due to rapid 
voltage collapse or rise, oscillograms and voltage distribution curves being 
given which support this view-point. Some remarks are then made on non- 
resonating transformers and also the impracticability due to the different 
characteristics of air and solid insulation of satisfactorily employing rod 
gaps for insulation co-ordinating purposes. 
Finally, some aspects of the operation of various wave front flattening 
devices are dealt with, these including the consideration of short cables, 
where it is shown that a flashover at a junction of the line and cable gives 
rise to H.F. oscillations. 
Dr. L. G. A. Srms.—sSpecification of magnetic qualities with particular 
reference to incremental magnetisation. The need for agreement (12.0). 
The established magnetic testing specifications of Britain, France, 
Germany and America are first examined and discussed. It is noted that 
they are all directed towards the testing of iron which is to operate either 
under steady-state conditions, as in D.C. work, or under alternating 
magnetisation, as in A.C. work. The case of magnetisation by combined 
A.C. and D.C., which has nowadays considerable commercial importance, 
is not included. The paper shows that this case offers difficulties. In 
particular, correlation of A.C. and ballistic test methods of measuring 
incremental permeability is not generally possible owing to unsymmetrical 
waveform distortion introduced by even harmonics. Certain aspects of 
this problem are demonstrated and suggestions for a common method of 
testing are put forward. But the views of other experimenters are sought. 
An agreement is already very desirable in the interests of present research 
work, the value of which should be enhanced by established methods of 
measurement. 
The view-point adopted throughout the paper is that the simple ballistic 
method of measuring permeability is only justified if the results are a guide 
to the behaviour of the: iron when excited by A.C. It is further assumed 
that the alternating flux conditions, which are typical of most practical 
applications of the iron, should be those specified for the measurement of 
iron losses. 
Although the paper is chiefly concerned with the effect of waveform upon 
