This paper provides evidence on the mechanisms influencing the patent output of a sample of small and large, entrepreneurial and established biotechnology firms from the input of indirect knowledge acquired from capital expenditures and direct knowledge from in-house R&D. Statistical models of counts are used to analyse the relationship between patent applications and R&D investment and capital expenditures. It focuses on biotechnology in the period 2002---2007 and is based on a unique data set drawn from various sources including the EU Industrial R&D Investment Scoreboard, the European Patent Office (EPO), the US Patent and Trademark Office, and the World Intellectual Property Organisation. The statistical models employed in the paper are Poisson distribution generalisations with the actual distribution of patent counts fitting the negative binomial distribution and gamma distribution very well. Findings support the idea that capital expenditures--taken as equivalent to technical change embodied in new machinery and capital equipment--may also play a crucial role in the development of new patentable items from scientific companies. For EPO patents, this role appears even more important than that played by R&D investment. The overall picture emerging from our analysis of the determinants of patenting in biotechnology is that the innovation process involves a well balanced combination of inputs from both R&D and new machinery and capital equipment.