Brood Production 



A multiple regression screen of brood production as a 

 function of phloem thickness and gallery density showed 

 gallery density (GD) accounted for a larger amount of 

 variance (30 percent) in brood production than did 

 phloem thickness (PT) (20 percent) (table 2). In a second 

 screen, substituting attack density (AD) for GD, PT 

 accounted for a larger amount of variance (20 percent) 

 than AD (13 percent) (table 2). The PT effect in each 

 case was clearly linear, while both GD and AD were 

 quadratic in nature. Each regression was highly signifi- 

 cant (P < 0.005). With this information as a base, data 

 were partitioned in each case to examine more closely 

 the nature of the generally quadratic effects of GD and 

 AD. The results are shown in figures 1 and 2. Brood 

 production was significantly and positively related to 

 egg gallery density and phloem thickness (R^ = 0.55; 

 P < 0.01). For example, at 0.3 m of egg gallery per 

 930 cm^, average brood production was 10 for thin 

 phloem, 23 for medium, and 33 for thick (fig. 1). At 

 2.1 m of egg gallery per 930 cm^ of bark, brood produc- 

 tion was 27 for thin, 56 for medium, and 75 for thick. 

 Brood production reached an asymptote at about 1.5 m 

 of egg gallery per 930 cm^ with no drop in brood produc- 

 tion out to 2.1 m of gallery per 930 cm^ of bark, the 

 highest egg gallery density observed in the test. 



Brood production was significantly related to PT and 

 AD, the combined effect being similar to that of PT and 

 GD (R2 = 0.38; P < 0.01). Brood production for seven 

 attacks per 930 cm^ was 24 in thin phloem, 45 in 

 mediiun, and 70 in thick. Curves flattened considerably 



a 

 o 



o cc 



a. < 



2.1 1.8 1.5 1.2 0.9 0.6 0.3 



EGG GALLERY (M) PER 930 Cm' OF BARK 



Figure 1 — Mountain pine beetle brood production per unit 

 area of ponderosa pine bark Increases with both egg gal- 

 lery density and phloem thickness. 



BP/930 cm2 = 0.9544 [UI - ^ (7 - 3.281 GD)" ] 



UI = (836.766) (0.03937 PT)^-2 



= upper intercept 

 n = 2.92 + 83.87 (0.03937 PJf -^ 



where 



PT = phloem thickness in millimeters 



GD = egg gallery density in meters per 930 cm^ 



P < 0.01 



R2 = 0.55 



Limits: < GD < 2.1. 



Table 2— Multiple regression screen of mountain pine beetle brood produc- 

 tion as a function of ponderosa pine phloem thickness (PT) and 

 egg gallery density (GD), and ponderosa pine phloem thickness 

 (PT) and attack density (AD) 



Phloem thickness ■ 



gallery density 



Phloem thickness 



attack density 



Variables 



R2 



Variables 





R2 



GD 



0.30 



AD 





0.13 



GD2 



.23 



AD2 





.08 



PT 



.20 



PT 





.20 



PT2 



.18 



PT2 





.18 



GD X GD2 



.35 



AD X AD2 





.20 



GD X PT 



.49 



AD X PT 





.35 



GD X PT2 



.47 



AD X PT2 





.33 



GD2 X PT 



.40 



AD2 X PT 





.30 



GD2 X PT2 



.38 



AD2 X PT2 





.28 



PT X PT2 



.48 



PT X PT2 





.23 



GD X GD2 X PT 



.55 



AD X AD2 X PT 





.40 



GD X GD2 X PT2 



.53 



AD X AD2 X PT2 





.38 



GD X PT X PT2 



.50 



AD X PT X PT2 





.36 



GD2 X PT X PT2 



.56 



AD2 X PT X PT2 





.31 



GD X GD2 X PT 



X PT2 .56 



AD X AD2 X PT 



X PT2 



.41 



4 



