A30 



the spring. However, boll weevils installed in cages between Oct. 15 and 

 Nov. 3 hi.d higher survival rates than weevils installed earlier or later. 

 Peak boll weevil emergence occurred in mid-May. 



842. . 1972. Photoperiodic sensitivity in the ontogeny of the boll weevil. 

 Environ. Entomol. 1: 568-571. 



Exposure of parent boll weevils to several photoperiods had no measurable in- 

 fluence on the incidence of diapause in the offspring. Photoperiod was largely 

 unable to override dietary influences on the regulation of boll weevil diapause. 



843. . 1974. Sampling problems with cotton insects. IPM, The Principles, 

 Strategies, and Tactics of Pest Population Regulation and Control in 

 Crop Ecosystems. Proceed. Workshop on Evaluation of role of Predators 

 in Crop Ecosystems, pp 22-23. Intl. Cts. for Bio. Con., U. of California, 

 Berkeley and Riverside. 



In our research sampling we think taking many single plant samples in a strati- 

 fied random technique has definite advantages. The greater the number of samples 

 we take the lower the standard error becomes. Also, by stratifying samples 

 across a field minor changes in plant growth patterns and arthropod dispersion 

 patterns can frequently be detected. Advantages of sequential sampling are its 

 time-saving features and predetermined probabilities of correctly assessing the 

 population.*" These probabilities are usually set at 80 to 90%. The stratified- 

 random research sampling technique using both whole-plant and D-Vac sampling 

 is very time-consuming and expensive. Since there are many small samples the 

 problem of data management becomes very difficult. We have largely solved 

 this problem by adopting IBM Port-A-Punch cards for recording raw data in 

 the field or laboratory. Using this technique we can often have data analyzed 

 15 minutes after it comes in from the field. The primary limitation with 

 se(fuential sampling is that each sample must be recorded as it is taken. Thus, 



