372 C. F. M. SWYNNERTON. 



spring and early rains, as they would be sitting in the shade by the wat-er, 

 attended it might be at first by a merely reasonable number of tsetse, every 

 newcomer or passer-by w^ould bring mth him an accession of flies till their numbers 

 became unbearable. Some would go again with departures, more would stay, and 

 they tended to increase in numbers from day to day. At that time every one, 

 sitting or travelling, carries a leafy switch and uses it incessantly. The probability 

 that the great majority of the flies of these village concentrations are males was 

 suggested further by my failure to find puparia at Gundoda's. 



Leaf-fall in relation to Concentration. 



It would seem that in one of the three forms of concentration I have referred to, 

 leaf -fall must be a highly important factor. That there are other factors producing 

 concentration is however certain ; for on my first arrival in the morsitans country 

 in June, concentration at the vlei centres was already in full operation, yet leaf-fall 

 (the season being late) was not yet noticeable. The same w^ould apply to my visit 

 at the end of May and beginning of June in 1900, when also high concentration 

 was present. When I left on 1st August 1918, leaf-fall was in many places pro- 

 ducing a visible effect, and it is possible that this may have brought about the 

 apparent increase in the numbers of j)allidipes at the Kanyezi morsitans centre. 

 Otherwise things had not, to all appearance, been so very different in June from 

 what they were in August, WTiatever the factors that bring about these " primary " 

 concentrations, the possibility is suggested that measures against them might 

 not have to be confined to the very brief period of maximum leaf-fall. All would 

 really depend on the value of the " scattered " fly for the local survival of the species, 

 and I am inclined for the moment to suspect that this may be greater in the case 

 of pallidipes — everywhere seemingly a rather scattered fly — than in that of 

 morsitans. 



The Kanyezi centre was certainly, to judge from its puparia, a focus of morsitans 

 rather than pallidipes, and it is conceivable that these primary centres may be a vital 

 point in this fly's social organisation. 



The rural population of brevipalpis, as I saw it, was much heavier and more 

 evenly distributed than that of morsitans and therefore less easily distinguished 

 from its rather more populated patches, in the primary thickets. In a less generally 

 suitable area (such as the granite-gneiss actually is) or in a season (if such occurs) 

 which really stripped the Brachystegia w^ooding of the " Oblong," it would doubtless- 

 undergo leaf -fall concentration. 



XVIII. — The Kecent Outbreaks of Nag ana in Mossurise. 



The fly, driven from a large piece of deciduously wooded country by Umzila's 

 measures, never greatly abandoned its permanent haunts except on the Sabi and 

 in the cleared portion of Gogoyo's, the Mwangezi, etc., and within 7 or 8 years 

 after the relaxation of the measures in 1889 it was found re-occupying its old (and 

 its present) wet season quarters in so far as bush conditions and travelling facilities 

 then allowed. Since then, within my own observation, the wooding of the hea\dly 

 deciduous areas hes increased greatly, and areas close to the British border that 



