Eggs of the blowflyProtophormia spec. were separated into anterior and posterior fragments of varying sizes. The operations were carried out between oviposition and the blastoderm stage. The partial larvae produced by the fragments were scored for the cuticular pattern they had formed. 2. The cuticle of the 1st instar larva carries 11 denticle belts which correspond to the anterior borders of the thoracic and abdominal body segments. These are considered the elements of a linear longitudinal pattern which starts with the head region. 3. Egg fragments of the sizes studied did not produce the complete cuticular pattern. 4. If denticle belts were present on the partial larvae formed in egg fragments, these always included the corresponding terminal pattern element (no. 1 in anterior, no. 11 in posterior fragments). Bigger partial patterns from anterior fragments may have any belt up to no. 10 as their most posterior belt, posterior partial patterns may start anteriorly with any belt up to no. 1, i.e. behind the head region. 5. After fragmentation during early stages of development, all eggs fail to form some pattern elements. Fragmentation thus causes a gap in the pattern. Extent and position within the pattern of this gap depend on level and stage of fragmentation. 6. With increasing egg age (developmental stage) at fragmentation, the gap in the cuticular pattern becomes progressively smaller. Eggs fragmented during or after formation of the blastodermal cell walls as a rule form all pattern elements. 7. The progressive reduction of the gap in the cuticular pattern is due to formation of bigger sets of pattern elements inboth partner fragments. I.e. on the average an anterior or posterior fragment of given size will produce more pattern elements if separated from the rest of the egg at a later stage than if separated early. 8. In order to produce a given set of pattern elements, a fragment needs to be bigger on the average when separated early than when separated later on. This applies to both anteriorand posterior fragments of the fragmentation levels studied. 9. According to these results, the egg ofProtophormia cannot be considered a mosaic of determinants for the different pattern elements at oviposition. The developmental fate of at least the more equatorial egg regions appears to become specified epigenetically during the period between oviposition and blastoderm formation. 10. Once the egg has become subdivided into blastoderm cells, it reacts as a developmental mosaic with respect to the pattern studied. 11. Preliminary results inDrosophila are compatible with these conclusions. 12. The results are compared to those obtained from other insect groups, and formal models for their interpretation are discussed. Pattern specification by interaction of terminal egg regions can be considered the common denominator for a number of egg types. 13. The results demonstrate that formally comparable processes of pattern formation occur in different insect egg types at different stages of development.
