Homozygous embryos do not show myoblast fusion defects.

Df(2L)CadN^Δ14 homozygous embryos, in which both CadN and CadN2 are deleted, show similar central nervous system defects to CadN⁴⁰⁵ homozygotes.

A small fraction of Df(2L)CadN^Δ14 mutant flies show a slight disruption of the pretarsus-ta5 boundary in late third instar leg discs. There are no visible defects in the leg discs or in adult leg morphology.

Mosaic eyes containing Df(2L)CadN^Δ14 clones show misrotated ommatidia, symmetrical clusters and loss of photoreceptor cells. In third instar eye discs with Df(2L)CadN^Δ14 clones, the majority of mutant ommatidial clusters reach a 90^oC angle prematurely compared to wild-type neighbours.

Df(2L)CadN^Δ14 homozygous R cell clones in otherwise heterozygous animals show normal extension of the growth cones into the lamina but show abnormal extensions within the lamina plexus, when scored at the first developmental stage in which extension can be evaluated. In approximately 59% of cases, no extension occurs toward the target cartridge in the lamina plexus and in the remaining 41% only a few thin, highly branched filopodia can be observed. When single mutant R1 and R6 cell axons are generated in otherwise wild-type adult animals, 31% of Df(2L)CadN^Δ14 R cells reach their targets, meaning that axon extension failure is less frequent in adults. Every R-cell subtype shows the extension phenotype when mutant for Df(2L)CadN^Δ14.

Df(2L)CadN^Δ14 clones in the lamina neuron do not affect the differentiation, morphology or projections out of the lamina but do affect the layer-specific targeting of these neurons in the lamina. Cartridges within or bordering a patch of Df(2L)CadN^Δ14 homozygous neuron clones in the lamina are abnormal in size due to abnormal numbers of (wild-type) R-cell termini innervating the cartridge and frequently lack all photoreceptor innervation. There is a strong correlation between the target column genotype and axon extension as wild-type R4 axons are much less likely to extend normally to a target cartridge with two or more mutant lamina neurons than to cartridges with 1 or 0 mutant neurons. Conversely, there is no correlation between genotype and cartridge of origin as axons from cartridges with more than 2 mutant lamina neurons misproject at similar frequencies to those from cartridges with 1 or 0 mutant neurons. There is no correlation between a particular lamina neuron subtype and the axon extension defects seen; all subtypes appear to be equally affected by Df(2L)CadN^Δ14 clones.