The sclera and cornea are dense and fibrous connective tissues that form the outer coat of the eye, which act to support and protect the eye from the surrounding environments. These load-bearing connective tissues can dynamically interact with changing physiological conditions^{1, 2}. Under elevated intraocular pressure (IOP), the corneoscleral shell transfers tensile, compression and shear stresses on the optic nerve head, and may contribute to the deformation of optic nerve axons and neurodegeneration in the visual system in glaucoma^3-5. To date, limited non-invasive imaging techniques are available for determining the microstructural organization and compositions of the corneoscleral shell globally and quantitatively in different environments, which are critical for understanding the dynamics and the biomechanical and biochemical properties of the eye⁶. In this study, we hypothesized that T2-weighted MRI (T2WI), diffusion tensor MRI (DTI) and magnetization transfer MRI (MTI) can detect and differentiate microstructural and macromolecular changes of the sclera and cornea in three experiments: #1: stepwise changes in biomechanical IOP loading (mimicking ocular hypertension⁷); #2: cross-linking (mimicking aging conditions⁸ and corneoscleral stiffening treatment^{9, 10}); and #3: glycosaminoglycans cleavage (mimicking glaucoma and myopic conditions¹¹). In a fourth experiment, we demonstrated in vivo the feasibility of magic-angle enhancement in the human sclera for examining the fibrous microstructures in the eye using T2*-weighted MRI (T2*WI). Experiment #1: T2 mapping was performed to 12 fresh, unfixed ovine eyes undergoing anterior chamber perfusion in the 9.4-Tesla MRI scanner using a plastic cannula connected to a saline bag and a pressure transducer (Figure 1). Six eyes were loaded at 0-40mmHg by raising the saline bag at different heights, followed by unpressurization back to 0mmHg. Six other eyes were cannulated but kept at 0mmHg as an unloaded control. Experiments #2 and #3: 16 ovine eyes were dissected at the central globes to give 4 sclera and 4 cornea tissue sections per eye. The tissue sections were treated with various concentrations of glyceraldehyde for cross-linking (#2) and chondroitinase-ABC for glycosaminoglycans depletion (#3) at 37^oC for 12 hours followed by T2WI, DTI and MTI assessments. All ovine eye experiments were performed using a 9.4-Tesla Varian/Agilent MRI scanner with the following imaging parameters. i) T2 mapping: TR/TE=1000/9.48ms, EST=9.48ms, number of echoes=5, in-plane resolution=130×130µm², slice thickness=1mm; ii) DTI: 2 non-diffusion-weighted images and 12 diffusion gradient directions at b=1000s/mm², δ/Δ=5/17ms, TR/TE=2300/27.8ms; iii) MTI: 9.5µT saturation pulses at 6000Hz off-resonance and 150ms pulse length, TR/TE=1500/8.43ms. Both DTI and MTI shared the same slice geometry, with in-plane resolution=140×140µm² and slice thickness=1mm. Experiment #4: A healthy adult subject fixated his eyes at 3 orientations (up, center, bottom) in the 3-Tesla Siemens Trio MRI scanner, and sagittal T2* mapping was performed to the right eye (IOP=11mmHg) using a 2D FLASH sequence, with TR/TE=50/3.52ms, EST=3.17ms, number of echoes=5, in-plane resolution=375×375µm², slice thickness=2mm and scanning duration=3min.In the ovine experiments, T2 in sclera and cornea increased non-linearly with biomechanical IOP loading at 0-40mmHg and remained higher than unloaded tissues after being unpressurized (Figure 2). Increasing dosages of glyceraldehyde (Figure 3) and chondroitinase-ABC treatments (Figure 4) decreased diffusivities and increased magnetization transfer in cornea. Glyceraldehyde also increased magnetization transfer in sclera (Figure 3). In the human experiment, T2* of the posterior sclera increased with increasing angle to the main magnetic field from 0^o to 55^o (the magic angle) as the subject fixated at different orientations (Figure 5). The T2 increase in IOP-loaded fresh ocular tissues might be explained by the diminishing spin dephasing between increasingly distant neighboring protons during collagen fiber straightening^{6, 12}. More importantly, the non-linearity of T2 changes echoed with recent studies showing non-linear uncrimping of collagen fibrils in the corneoscleral shell with IOP loading^{13, 14}. T2 of IOP-loaded tissues remained high after being unpressurized suggestive of MRI detection of hysteresis or hydraulic effects^{15, 16}. The increasing magnetization transfer and reducing diffusivity in the glyceraldehyde-treated ocular tissues suggested that MTI and DTI could characterize the state of collagen cross-linking such as increasing cross-linking packing density¹⁷ and reducing permeability¹⁸ in the eye under different physiological conditions. DTI and MTI may also detect glycosaminoglycan removal which is often a cause of altered creep and stiffness in glaucoma and myopic eyes¹⁹. As our multi-modal imaging techniques share similar physical principles in both high-field MRI and clinical MRI, our initial magic angle-enhanced MRI demonstration to detect the fibrous tissues in the human eye opened up the possibility for future non-invasive multi-modal MRI assessments of the corneoscleral biomechanics in the living human eyes, which may help identify new modifiable risk factors and guide vision preservation in a variety of ophthalmic diseases.