Evolution has driven genetic diversity of life on Earth, but also created highly complex genomes that are difficult to sequence. Current draft genomes can have thousands to hundreds of thousands of contigs rather than chromosomes, containing incorrect assemblies, gaps and errors. With rapid advances in long-read technologies, it is becoming possible to resolve complex genomes, including repetitive, polyploid plant genomes. Despite the technology being available, a challenge persists: the extraction of pure high molecular weight DNA suitable for long-read sequencing. This is particularly true of recalcitrant native Australian trees such as Eucalypts and Acacias. To resolve this, firstly we optimised a density gradient based nuclei extraction to remove cytoplasmic secondary metabolites, phenols and limit reads from high copy count plastid genomes. Secondly, we optimised a gentle high-molecular weight DNA extraction free of columns and high centrifugation, to limit DNA fragmentation. Finally, DNA was purified and size selected by gel electrophoresis. For sequencing, we adopted the portable MinION sequencer from Oxford Nanopore Technologies. Using these approaches, we can approximately obtain over 10 gigabases of sequencing from a single MinION revC flow cell and over 15 gigabases with new revD flow cells, up to 23 gigabases. This includes quality reads over 200 kb in length, average N50 values over 20 kb and some N50 values exceeding 45 kb. Such ultra-long reads assist the assembly of high quality genomes, from telomere to telomere.