E.O.JOHNSON AND L. MALTER This is not the case for the SPM. The probe or elec- trode potentials for this case can affect the plasma Ya tanh X potential in such a way as to alter the quantities being measured. If a hot cathode is present, then the use of a positive probe may give rise to oscillations during the afterglow period. This is an undesirable state of affairs the measuring device affects the quantity it is designed to measure. If a hot cathode or other copious electron source is not t(as in the case of a co 30 cathode discharge, or for the isolated anode-grid region plasma of case 3 above), then it is possible to use a single probe combined with the other electrodes as a IX. ANALYTICAL EXPRESSIONS FOR id aNd i Let us consider a symmetrical double probe system in which:(1)the probes are identical, (2) the randor current density at both probes is equal, (3)the contact FIG. 15. Ideal double probe current-voltage characteristic mf between the probes is potential near both probes is the same. Then from Eq current density at the probe to the random current density in the plasma near the probe(when the probe tential is v jd/0=e(mkT)=e-(1.619)0=0.0021 From Fig. 11 we see that the measurements extend to points where the current to either probe becomes about double the above value. Thus the maximum value assumed by ja/jo in this case is about 0.005. Thus less Y than the top 1 percent of the electrons are sampled in this case In other tubes than those described here, up to 5 ercent of the electrons were sampled by means of probes with area ratios of 25. In these cases, too, the results always corresponded to Maxwellian distribu tions. It may obviously represent a dangerous extra polation to conclude from the properties of this small Y:In IR sample that the electron velocity distribution is com pletely Maxwellian VIII. FURTHER CONSIDERATIONS OF THE SPM AND THE DPM It may appear that the single probe can be employe i gthe pfasbioas of tt doubl e perot doubles dering the Fic. 10. Theoretical and ie prime mel hoset on arent plot o and the remainder of the system as constituting the other probe of the pair. This is actually possible enly in (8)o=1. Then from Eq(6) ertain special cases. From the earlier description of the DPM we see For this symmetrical case, ip" Ip:=Ip. Then: Sip=2t py potential of the probes)the space potential in the where ip is the positive ion current to either probe plasma is unaffected by changes in va. The plasma is Making use of this relation and the fact that ie, ia +ip, dominant in determining its own potential as well as we can transform Eq.(18)into that of the floating probes. The potential of the latter id/ip=tanh(φVd/2). is determined by that of the plasma and the condition A plot of this function is given in Fig. 15. It is seen to of equality of electron and ion currents to the probe have just the shape of the observed data. ( See Figs. 11 and 13