Harmful supernumerary chromosomes in a mealy bug population.

HE supernumeraries, or B chromosomes, of the mealy bug, Pseudococcus obscurus, have been shown to undergo preferential segregation both in oogenesis and in spermatogenesis (NUR 1962a). In a female with two supernumeraries, they do not pair during the first meiotic division but usually pair and segregate regularly in the second division. However, when one or an odd number is present the unpaired chromosome segregates more often into the second polar body than into the egg. As in other species with the lecanoid chromosome system (HUGHES-SCHRADER 1948; BROWN and NELSON-REES 1961), the paternal set of chromosomes of the males becomes heterochromatic in the embryo and remains so throughout development. Meiosis consists of two divisions during which homologous chromosomes do not pair. The first division is mitotic-like, and the secondary spermatocytes contain a diploid number of chromosomes. During the second division, the euchromatic and heterochromatic sets segregate to opposite poles, and, of the four products of meiosis, only the two euchromatic derivatives form functional sperm. The supernumeraries in the male are positively heteropycnotic up until meiosis but then become negatively heteropycnotic. During the first meiotic division, they also divide equationally, but during the second division, they usually segregate preferentially, with the euchromatic set. Thus, an unreduced number of supernumeraries is usually included in the sperm, and the preferential segregation leads to the creation of an accumulation mechanism. The efficiency of the preferential segregation of the supernumeraries is quite high so that, in a male with one supernumerary, from 80% to 100% of the sperms carry a supernumerary. The accumulation of the supernumeraries in the males is greater than their elimination during oogenesis; therefore, the frequency of the supernumeraries would be expected to increase unless this increase were counteracted by selection against at least some of the individuals with supernumeraries. Sampling of several wild populations of this species in California over a few years indicated that the frequency of the supernumeraries remained about the same (NUR 1962b). The frequency of the Supernumeraries varied greatly between populations and was correlated with geographical distribution but not with