Axons are up-to-a-meter-long cable-like cellular processes of neurons. The proper function of nervous systems requires that axons grow and wire up correctly during development or regeneration. The uniquely challenging architecture of axons has to be sustained for an organism's lifetime, and renders them key lesion sites during healthy ageing, in injury and neurodegenerative diseases. Notably, axon degeneration is considered as the cause rather than consequence for neuron decay in the context of various neurodegenerative diseases. The structural backbones of axons are formed by parallel bundles of microtubules (MTs) which also provide the highways for life-sustaining long-distance transport between cell bodies and the growth cones or synaptic endings. To better understand axon development, regeneration, maintenance and degeneration during ageing, my PhD project has focused on mechanisms underpinning the regulation of MT bundles in axons. For this, I have capitalised on fast and genetically and experimentally amenable research possible in Drosophila neurons, both in primary culture and in vivo. I have used systematic combinatorial genetics and pharmacological approaches to unravel mechanisms and roles of actin as well as the cortical collapse factor Efa6 in MT regulation during axon formation and maintenance. I was able to gain a number of novel mechanisms contributing to the de novo alignment and maintenance of ordered MT bundles. First, it has been proposed that Spectraplakins (large actin-microtubule linkers) guide the extension of polymerising MTs along cortical F-actin, thus directly laying axonal MTs out into parallel bundles. Here, I have used manipulations of actin networks as well as hybrid constructs of Shot where the actin binding domain was replaced by actin associating domains of other molecules. My data strongly suggest that Shot's ABD domain has unique properties that can sense specific properties of F-actin networks, and this is important for its ability to appropriately regulate MT behaviours. Second, using combinations of actin and Shot manipulations, I found that Shot displays not only these actin-dependent guidance functions, but it displays novel actin-independent function in MT bundle maintenance for which I present a working hypothesis. Third, I found a novel and Shot-independent role of axonal actin in maintaining MTs and promoting axon growth, and my results suggest that these functions involve promotion of MT polymerisation. MT maintenance is therefore mediated through two complementary mechanisms involving Shot on the one hand and actin on the other, and simultaneous removal of Shot and actin leads to entire loss of axons. Finally, I have unravelled novel axonal functions of the cortical collapse factor Efa6 which serves as a check point in MT bundle maintenance by eliminating "off track" MTs that have escaped the axonal bundle organisation. In the absence of this factor, a gradual increase of disorganised, criss-crossed MTs occurs as a matter of days. These new mechanisms strongly suggest that different MT-regulatory mechanisms act in parallel in axons and complement each other in one common mechanism of MT bundle formation and maintenance. I propose here a local homeostasis model of axonal MT bundle maintenance which provides new ways to think about problems of ageing as well as a range of different neurodegenerative diseases.