Due to the small size of the sample it is difficult to grind sample in a mortar & pestle and direct grinding in eppendorf also does not gives good results.
Although I have not isolated DNA from mosquito, but I have worked with small and hard cancer tissues.. I used to put the small tissue sample on sterile glass slide and finely chop the tissues using sterile surgical blades (blade size no. 23 or 22 is fine) and follow with SDS-proteinase k digestion. all the best!!
This work pretty fine: "Each mosquito pool was triturated in 500 mL of cell culture medium (high glucose Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) with 10% heat-inactivated fetal bovine serum, 100 U/mL penicillin, 100 mg/mL streptomycin, and 2.5 mg/mL amphotericin B using two stainless steel beads (5 mm; Qiagen) in a TissueLyser (Qiagen) for 2 min at 50 oscillation/s. " from this publication: https://www.researchgate.net/publication/256614069_First_Nationwide_Surveillance_of_Culex_pipiens_Complex_and_Culex_torrentium_Mosquitoes_Demonstrated_the_Presence_of_Culex_pipiens_Biotype_pipiensmolestus_Hybrids_in_Germany
Article First Nationwide Surveillance of Culex pipiens Complex and C...
Surely you'll get not much DNA, therefore the extraction has to be as efficient as possible. The Eppendorf cup with a fitting teflon pistil is usually a good idea, if you have many samples and it should be easy to do and cheap. But there are also glass vials with ground glass surfaces and fitting glass pistills, which also improve grinding.
Did you use liquid nitrogen to increase the grinding efficacy? Also longer incubation times (overnight) in lysis buffer containing (among other chemicals) SDS, proteinase K at 55°C increases yield.
And finally also the DNA extraction method after the grinding is important to reduce loss (at the moment I'd prefer a kit to do this as it seems imo superior to other methods). If you have a good protocol you can adapt it by e.g. reducing the washing steps to have less loss.
If you have access to a bead beater machine of some description eg: Mp Bio Fastprep I would recommend using this. You can homogenise directly in the lysis buffer from your DNA extraction process.
i had used half of a larva for DNA extraction and then used that DNA to amplify 13 micorsatellite loci and all of the samples work perfectly. i had used simple ammonium accetate method for DNA extraction. so if you are carrying genetic studies then even a single leg of mosquito is enough to get DNA. if you are detecting a virus of protozoan then you go for pools of mosquitoes as you have large number of collection to deal with.