618 THE MONTHLY BULLETIN". 



Soon after hatching' begins the surface of the mother scales becomes 

 dry and light brown in color, and begins to adhere less closely to the 

 twig. At this stage tliov are found almost wholly on the smaller twigs, 

 being arranged in places in an imbricated manner, covering the twig- 

 like fish-scales. When the young are all out, the dead shells are very 

 apt to drop off, leaving whitish marks. One does not find many old 

 shells of the preceding year, as is the case with the black scale. 



This scale throws out a great deal of honeydew, the resulting smut 

 making the trees very dirty, but this takes place somewhat later in the 

 season than is the case with the black scale, so that the navel oranges 

 have usually been picked ; but the Valencias become much soiled. 



With regard to the following experiments, in all ca^es the trees 

 selected were the most badly infested that could be found of a suitable 

 size. Sodium cyanide and the Woglum system were nsel. 



As regards the condition of the scale when fumigated, young were still 

 being produced under the old shells up to August 21st. On trees fumi- 

 gated before that date apparently all the young under the shells at time 

 of fumigation were killed, but the old mother-shells were not affected, 

 so that a few young would be born after fumigation at any date prior to 

 August 21st. 



On or about July 1st, this year, I took a count, on each tree, of the 

 adult scale that I could find in a ten-minutes' search. The number of 

 adults that could be counted in the same time on a neighboring tree was 

 also taken as a check and indication of the condition of the other trees 

 in the same grove. But on the trees fumigated the number found in 

 ten minutes would be about all the adults that were on the tree, whereas 

 on the other trees this was by no means the ease. 



The results are not quite uniform, and the number of trees treated 

 was not large, but I think the results may be held to indicate that 

 there is a good chance of obtaining a fair killing on this scale between, 

 say, July 20th and end of August. Between these dates there does not 

 seem to be any marked difference in results. With regard to dosage, it 

 seems to be that a three-quarters schedule would be advisable, to give a 

 safe margin, while a heavier dosage than that does not seem to be 

 justified. 



With regard to the work of the contractors last fall, all the fumiga- 

 tions done after the middle of September were failures as far as the 

 "longulus" is concerned. One grove fumigated on September 7th with 

 an ' ' estimated ' ' schedule was also a bad failure, but in this case a very 

 poor result was also obtained on the black scale. In one grove a very 

 good result, both on the black and "longulus," was obtained on Septem- 

 ber 13th, but the same outfit got very poor results on "longulus" in 

 another grove, close by, on September 17th, though they again got a 

 good killing on the black. In both cases a five-eighths schedule was 

 used. In another grove, where we got a very satisfactory result on 

 "longulus" on September 4th on the six trees we fumigated, the con- 

 tractor got a very poor result on it over the rest of the grove when it 

 was fumigated only twelve days later, September 16th, using a five- 

 eighths schedule, though he killed the black very well. 



I may mention that all of the 57 trees fumigated experimentally are 

 now fairly free from black scale as well as ' ' longulus, ' ' though in some 

 of these groves there is much black scale on the surrounding trees. This 



