346 H. I. JENSEN. 



also tend to raise the spots on which they stand by catch- 

 ing atmospheric dust and debris. In this way 'melonhole* 

 country often forms in coastal districts. Its formation 

 may be aided by the alternate expansion and contraction 

 of the soil on wetting. Indeed coastal 'melonhole' country 

 is usually on peaty clay soils and the knobs being more 

 peaty than the hollows would probably expand at a more 

 rapid rate and to a greater extent on 'wetting.' 



While such causes might have helped to produce the 

 beginnings of gilgai country in the Pilliga Scrub, they are 

 no longer operative, for the soils are not peaty in nature, 

 nor do the roots of the plants existing there at present act 

 as do the tussock grasses of swamps. 



The roots of belar and brigalow being mainly surface 

 roots certainly help to maintain the hummocks intact, but 

 I cannot believe that they have produced these irregu- 

 larities. Indeed belar grows abundantly on country which 

 is not 'gilgai,' and I believe this is the case with brigalow 

 also. 



6. Mr. J. F. Campbell, l.s., in a paper on Soil Physics 

 read at a meeting of the Institute of Surveyors on May 18th 

 1909, suggested that 'melonholes' ('crabholes' or 'small 

 gilgais') are due to the effect of sodium carbonate in 

 destroying soil crumbs and causing the soil to subside by 

 counteracting the cementing crumb-producing properties 

 of carbonate of lime. 



Possibly this cause may at times produce 'melonhole' 

 country, but I cannot believe that our gilgai country could 

 have been produced in this way. Chemical evidence is 

 also against this theory, for in my soils the hummock soils 

 had the highest alkalinity, whereas their alkalinity should 

 be least under Mr. Campbell's hypothesis. 



7. Mudsprings exist according to the statements of many 

 old pioneers of the Pilliga Scrub in the country lying about 



