Analysis of Daily Growth Patterns in Young-of-Year Male Dwarf Surfperch (Embiotocidae: Micrometrus minimus) Suggests Alternative Tactics: Breed or Grow after Birth

Discontinuous variation in daily growth patterns suggests that there are alternative life-history tactics among males of a coastal fish species. Male dwarf perch are known to be mature and reproductively active at birth, exhibiting a high level of material and behavioral expenditure on breeding early in life. Males experience a substantial cost of reproduction in reduced early growth and survival relative to females. Through analysis of daily growth increments in otoliths, we showed that young-of-year males from two populations exhibit a polymorphism in daily growth patterns. A distinctive episode of narrow growth increments and slow growth punctuated the summertime growth record of some males, whereas increment widths were more uniform and growth was more rapid in other males. Discriminant function analysis revealed that features of the daily growth record varied discontinuously, falling into two distinct groups. Early-born males were more likely to exhibit the slow growth pattern, and late-born males were more likely to show the fast growth pattern. We interpret these results as indicating that late-born males, with relatively few opportunities for mating, defer reproductive effort and minimize the growth cost of reproduction. There is likely to be a strong selective advantage for early-season birth dates among males in these populations.

Fig. 1.

Summary variables used in analysis of daily increment width patterns. Open and closed dots represent smoothed (four-day running average) increment width plotted against increment number (= age in days from birth) for a selected individual. The line represents the bivariate regression of increment width against age and is used to identify increments that are narrower than expected (open dots). The large triangle identifies the narrowest increment and defines the minimum increment width (MIW) and the age that the MIW occurred (MIW age). The sequence of narrow increment (SNI; see text) is represented as well. The date that the MIW occurred (MIW date; not shown) is equal to birth date + MIW age

Fig. 2.

Otolith daily increment width patterns. Smoothed increment width is plotted against increment number. Each individual record is aligned with MIW age at 0 for clarity. Records are grouped by growth type/sex; male growth type was determined subjectively

Fig. 3.

Summary variables of otolith daily increment width patterns. Four summary variables are plotted for each individual, with sex/growth type indicated by the symbol (see legend). Male growth type was determined subjectively. The open symbols represent duplicated values because of a second occurrence of the MIW (four individuals in the TB population). (A), (C), and (E): TB population. (B), (D), and (F) MB population

Fig. 4.

Growth rate of females and two classes of males over the first growing season, TB population 1986. Growth rate (± SE), estimated from bivariate regression of length versus age, is plotted against birth date, divided into biweekly cohorts. Datapoint labels represent the sample size. Sex and growth type are indicated by the symbol (see legend); male growth type was determined subjectively. Points for some cohorts are missing because of low sample size (fewer than 3)

Fig. 5.

Histogram of birth date distribution (as the proportion within each birth week) of males, by growth type. (A) TB population. n = 138 fast males, 145 slow males. (B) MB population. n = 11 fast males, 14 slow males

Accepted: 28 Jun 2000