This document describes the design, analysis, construction and testing of KiwiSpec, a fibre-fed, high resolution astronomical spectrograph of an asymmetric white pupil design. The instrument employs an R4, 31.6 groove mm⁻¹ échelle grating for primary dispersion and a 725 lines mm⁻¹ volume phase holographic (VPH) based grism for cross-dispersion. Two versions of the prototype were designed and constructed: an 'in-air' prototype, and a prototype featuring a vacuum chamber (to increase the stability of the instrument). The KiwiSpec optical design is introduced, as well as a description of the theory behind a cross-dispersed échelle spectrograph. The results of tolerancing the optical design are reported for alignment, optical fabrication, and optical surface quality groups of parameters. The optical windows of an iodine cell are also toleranced. The opto-mechanical mounts of both prototypes are described in detail, as is the design of the vacuum chamber system. Given the goal of 1 m/s radial velocity stability, analyses were undertaken to determine the allowable amount of movement of the vacuum windows, and to determine the allowable changes in temperature and pressure within and outside of the vacuum chamber. The spectral efficiency of the instrument was estimated through a predictive model; this was calculated for the as-built instrument and also for an instrument with ideal, high-efficiency coatings. Measurements of the spectral efficiency of various components of the instrument are reported, as well as a description of the measurement system developed to test the efficiency of VPH gratings. On-sky efficiency measurements from use of KiwiSpec on the 1-m McLellan telescope at Mt John University Observatory are reported. Two possible exposure meter locations are explored via an efficiency model, and also through the measurement of the zero-order reflectivity of the échelle grating. Various stability aspects of the design are investigated. These include the stability of the optical mounts with temperature changes, and also the effect of the expansion and contraction of the supporting optical tables. As well, the stability of the in-air prototype was determined through measurement of the movement of thorium-argon emission lines within spectra as the temperature, atmospheric pressure and relative humidity (naturally) varied. Current and planned testing for determining the stability of the vacuum chamber prototype is discussed.