Understanding how transcriptomes differ within and between species is key to understanding everything from personalized medicine to phenotypic changes on evolutionary timescales. Hawaiian flies provide an exceptional model for understanding population structure as well as sympatric and allopatric speciation. >500 species of Drosophila unique to the Hawaiian archipelago have been described. These islands are located thousands of kilometers from continents and it is believed that all the species in this extensive radiation developed from a single founder female approximately 25 million years ago. The individual volcanic islands arise in series as the pacific plate passes over a mantle hotspot, such that the geological age of each is well established, placing firm limits on when new founder populations arrived. In addition to reproductive isolation events due to initial founder effects, there are subsequent isolation events because of subsiding volcanoes and sea level rise that fragment old islands, and lava flows that fragment ecosystems on young islands. There are major microclimate differences due to island features such as altitude and rain shadows that promote adaptive radiation. Mating behavior is particularly striking as males perform elaborate displays and bouts in leks. There are also morphological specializations for male-male competitions (such as battering ram heads in D. heteroneura). At abstract submission, we have performed mRNA sequencing for a total of 149 samples (over 1.5 billion RNA-seq reads) to explore expression variance due to sex (N=2), tissue (N=8), strain or species (N=10), and island (N=4). We are assembling transcripts de novo and have mapped reads to the D. grimshawi genome (the only currently assembled hawaiian genome). Fortunately, >84% of our reads from D. silvestris, D. hemipeza, D. heteroneura, and D. hawaiiensis uniquely align to the D. grimshawi genome. We are also reannotating the D. grimshawi genome and evaluating the SNPs and Indels in transcriptome for within- and between-species variance in addition to expression level variance. Variance in male- and particularly testis-biased (sampled separately from male reproductive tract) gene expression is marked. These observations highlight the importance of sexual selection in speciation of these flies (the Kaneshiro hypothesis) at the genome deployment level, despite the many island niche features that one would expect to drive selection for individual fitness. These results indicate that this genomic model will be a powerful tool for exploring population and evolutionary questions.