Males from a wild-type laboratory stock (M56i, Amherst), were mated with virgin females of the M-6 stock, In(1)sc 8 In(1)dl-49, y 31d sc 8 w a v Of f , and 149 spontaneous independent non-mosaically transmitted, as well as 8 incidentally detected, mosaically transmitted, X-linked recessive lethal mutations were recovered from 95704 F 2 cultures. 152 mutations were mapped over the entire length of the X-chromosome by complementation and/or crossover tests. Altogether 70 mutations, but no more than 63 mutations with lethal effects supposedly localized to single vital loci, fell into the 5 covered regions comprising about one-third of the X-chromosome that were studied in Lefevre's (1981) analysis of X-ray-induced lethal mutations. The distributions of spontaneous and X-ray mutations among the 5 regions were similar. Although there were far too few spontaneous mutations to make a meaningful comparison of relative mutability on a locus-by-locus basis, those loci displaying a relatively higher X-ray mutability, when taken as a group, tend to display a relatively higher spontaneous mutability, and those loci displaying a relatively lower X-ray mutability, when taken as a group, tend to display a relatively lower spontaneous mutability. Therefore, the data from lethal mutations are in agreement with the data from specific locus visible mutations and support the conclusion of Shukla et al. (1979) in being consistent with the assumption of proportionality between spontaneous and X-ray-induced mutation rates. Genetic analysis revealed that a minimum of 7.2% ( 11 152 ) of the were associated with chromosomal rearrangements. Within all covered regions, there were some 94 mutations that were tested for allelism, but only one possible case was found of a locus, l(1) L13, with as many as 3 independent intragenic mutations; it is unlikely that among the 152 mutations there were any other such loci. The locus most frequently involved in any kind of lethal mutation was the l(1) J1 locus near the distal tip of the X-chromosome with 5 mutations that included 3 ostensible multi-locus deficiencies. At equivalent entire X-chromosome recessive lethal mutation rates, the estimated per locus spontaneous mutation rate for non-mosaically transmitted X-chromosome lethals is about the same, or not lower than by a factor of 1 3 , as the estimated per locus spontaneous mutation rate for non-mosaically transmitted X-chromosome recessive visibles at 13 specific loci studies earlier by Schalet (1958, 1960) and re-evaluated here. In the earlier study of spontaneous visible mutations, at least 80% of the newly arisen non-clustered mutations (counting both transmitted and untransmitted), showed a mosaic distribution of mutant tissue by appearance and/or genetic test, but only 41% of the mutations that were germically completely mutant in transmission originally appeared in F 1 females as somatic mosaics. Upon reconsideration it is now appreciated that many newly arisen mutations would have been undetected because they were cryptic somatic mosaics. Accordingly, it is judged that the inclusion of an estimate of the numbers of undetected cryptic somatic mosaics would raise to at least 2 3 the proportion of newly arisen mutations that were somatic mosaics and germically complete in F 1 females. Among the 141 newly arisen non-clustered recessive lethal mutations that were germically complete, there were 9 mutations with dominant visible mutant effects and at least 3–5 of these were cryptic somatic or gonosomic mosaics. From the data on dominant and recessive visibles I infer that the majority of all spontaneous recessive lethal mutations were, although germinally complete, also mosaics of one type or another in F 1 females. Furthermore, if transposable elements were responsible for the mosaics observed or inferred to have occurred, then they must have been mobilized very early in cleavage in order to produce mutant cleavage nuclei more usually ancestral to somatic and germinal tissue or to somatic tissue alone, rather than germinal tissue alone.