The Impossible Partnership: Apple’s Coming Reckoning with China

The role of Apple’s chief executive still appears deceptively familiar: steward a portfolio of iconic products, manage a sprawling global supply chain, appease Donald Trump, and deliver outsized returns to shareholders. As Apple is poised to transition from Tim Cook to John Ternus, however, that conception no longer reflects the reality of what the position entails.¹ Ternus will inherit a position fundamentally altered by geopolitics, one that now sits uncomfortably closer to questions of national interest, regulatory authority, and public trust in ways few corporate roles ever have.

John Ternus is Apple’s fifty-year-old senior vice president of hardware engineering.² A twenty-five-year company veteran who earned Tim Cook’s trust by shepherding the Mac’s transition to Apple Silicon, Ternus will take over as CEO on September 1, 2026, while Cook will step into the role of executive chairman. Some inside the company reportedly consider Ternus too risk-averse, and in 2023, he publicly laughed off concerns about Apple being late to generative AI.³ Events, including Apple’s debacles with China and AI (to be discussed below), have proven him badly wrong.

The choices made by Cook with respect to the depth of Apple’s integration with China—and its responses to growing pushback from the United States—are forcing a reckoning that can no longer be avoided or hedged. The following essay is not so much about Ternus but about the structural factors that will shape and constrain his tenure.

The Derisking Illusion

Apple’s exposure to China and the directives set by the Chinese Communist Party (CCP) is often framed in Western policy and investor circles as a manufacturing problem that can be gradually “derisked” through diversification, by shifting production to India or Southeast Asia, or reshoring to the United States. That framing is no longer adequate. Today, Chinese authorities shape not only where Apple’s devices are assembled but the regulatory conditions under which Apple governs user data, structures its supplier ecosystem, and deploys core software features. These pressures are increasingly embedded in Apple’s operating model, constraining strategic choices in ways that Ternus will not be easily able to unwind through supply chain reshuffling or geographic expansion. What distinguishes Apple is how its China exposure reaches into functions normally treated as settled questions of corporate control.

Apple’s data governance in China is the clearest illustration. In mainland China, iCloud is operated by AIPO Cloud (Guizhou) Technology Co., Ltd., a provincial government-backed cloud services operator that functions within Guizhou’s big data SOE (state-owned enterprise) ecosystem. Guizhou Big Data Group, the provincial big data platform that reportedly serves as Apple’s iCloud partner in mainland China, anchors that ecosystem.⁴

In late 2025, Chinese provincial disciplinary authorities announced an investigation into Xu Hao, then chairman of Guizhou Big Data Group, for suspected violation of party discipline and national law.⁵ Apple’s services continued without disruption, and the company made no corresponding public disclosure. But the episode illustrated something that rarely surfaces in earnings calls or investor filings. Apple is not responsible for storing its Chinese user data; instead, a Chinese state-backed company fulfills this function. That company operates inside a system where CCP authorities can intervene directly, quietly, and without any formal public process.

Normally, a company like Apple would store user data through a private contractor bound by commercial agreements and answerable to courts. Guizhou Big Data Group is neither. It is a provincial state-owned enterprise controlled by the Guizhou government, the leadership of which serves at the discretion of party authorities. Under Chinese law, it is legally required to cooperate with national intelligence and security agencies when asked, and it has little legal basis to refuse. The encryption keys that protect Chinese iCloud data sit on servers inside facilities under the company’s control. When Chinese authorities want user data, they ask a Chinese government entity operating under Chinese law. Apple can and often does say the handover was not its decision, essentially disclaiming responsibility for the transfer of user data.⁶

Following Xu’s investigation, Guizhou Big Data Group was quietly rebranded as Guizhou Big Data Industry Group Co., Ltd. Chinese state-owned enterprises are routinely reorganized and renamed when leadership comes under scrutiny, a practice that is meant to signal institutional renewal while mostly changing nothing of substance. Apple almost certainly had no role in the restructuring and no apparent interest in drawing attention to it. Apple’s part, as with most things in this arrangement, was simply to say nothing and continue.

As Xi Jinping stated in 2020, China must tighten the dependence of international production chains on China, which means deploying powerful countermeasures and deterrent capabilities based on artificially cutting off supply to foreigners. The pandemic validated Xi’s logic from the other direction. When China’s initial lockdowns hit in early 2020, manufacturers worldwide—from pharmaceuticals and medical equipment to basic industrial goods—discovered just how exposed they were.⁷ The crisis landed on top of an already deteriorating bilateral relationship: under the first Trump administration’s tariff campaign, Washington spent two years trying to force supply chain separation, and the pandemic accelerated a reckoning that was already underway. China’s strict zero-Covid controls in 2022, including factory lockdowns, logistics disruptions, and sudden shutdowns, caused a second wave of economic damage for companies that had already survived the first.

By the end of 2023, China had fallen to the third-largest source of U.S. imports, for the first time since 2008.⁸ This was when Western governments reached for a new vocabulary to describe what they were doing: “derisking” rather than decoupling, a framing designed to signal selectivity rather than wholesale separation. Senior Biden administration officials, including Secretary of State Antony Blinken and Treasury Secretary Janet Yellen, visited Beijing in 2023 in an effort to resume engagement, even as both governments pursued supply-chain strategies that made deeper entanglement less likely.⁹

For Apple, the derisking frame never truly fit. Patrick McGee, the journalist and author of Apple in China, told the news outlet Rest of World in May 2025 that Apple “wants the perception that they are moving a lot to India” while maintaining “the reality of continuing to build as much as they can out of China.”¹⁰ He argues that an iPhone that says “Made in India” on the box will be no less dependent on the China-centric supply chain than any iPhone previously purchased. TechWire Asia also noted that Apple’s dependency on China now extends beyond final assembly to components, meaning even Indian-assembled iPhones rely heavily on Chinese parts and expertise.¹¹

Why There Is No Alternative to China

Apple releases new iPhone models on a fixed annual cycle, with the majority of sales concentrated in the first several weeks. Missing that window even by days means significant lost revenue; hitting it requires assembling tens of millions of devices in a matter of months, with every component, every production line, and every quality-check procedure running at full speed simultaneously. That kind of coordination depends on more than factories. Each new iPhone involves thousands of last-minute design adjustments.

In China, Apple engineers are embedded directly in facilities, problems get resolved overnight, and changes ripple quickly through a dense cluster of nearby suppliers. China’s manufacturing ecosystem was built around Apple over decades. Reporting by the Financial Times in 2023 found that at an iPhone factory in southern India run by Tata, one of India’s biggest industrial companies, roughly half of all components coming off the line were rejected—before the satisfactory ones could be sent to Foxconn, Apple’s primary manufacturing contractor in China, for assembly.¹² Chinese suppliers, former Apple engineers said, often completed tasks expected to take weeks in a matter of days. India’s manufacturing base has improved since 2023, but analysts tracking Apple’s supply chain note that yields and speed at Indian facilities still trail China’s.¹³

Foxconn’s India factories produce only about a fifth as many iPhones, achieve lower success rates on the production line than the 98 percent-plus yields in factories in the eastern Chinese city of Zhengzhou, and continue to struggle with basic issues such as infrastructure gaps, immature lines, hygiene problems, and heat-related power shortages.¹⁴

India does not just lag behind China on production; China can actively prevent it from catching up. In mid-2025, under quiet pressure from Beijing, Foxconn reportedly pulled over three hundred Chinese engineers from its India factories, cutting off the technical expertise needed for Apple’s production ramp-up.¹⁵ According to multiple sources cited by Bloomberg and Rest of World, Beijing’s encouragement to curb technology and labor transfers to India and Southeast Asia was delivered verbally to regulatory agencies and local governments, who in turn made those expectations clear to companies like Foxconn.¹⁶ The directive was then passed on to Foxconn staff verbally, and workers with visas and plane tickets to India were told to cancel their trips; those already stationed there were told to return home before an unspecified date.

The Foxconn episode and the iCloud arrangement share the same basic structure: Beijing does not need to issue formal orders to get what it wants from companies operating inside China’s system. That is the arrangement Ternus will inherit. It is also one that has grown harder to defend publicly. Shareholders, regulators, and journalists have become more attentive to what Apple does and does not disclose about its China exposure. When a senior official at the company storing Chinese users’ iCloud data comes under a CCP investigation, Apple says nothing. When Beijing quietly pressures a key manufacturing partner to withdraw hundreds of engineers, Apple says nothing. Apple discloses what it is legally required to disclose, but the gap between what is disclosed and what is materially relevant to investors has widened—and that gap is itself a liability.

The same problems, in varying degrees, afflict every alternative manufacturing host nation that analysts and investors typically point to as plausible replacements for China. The American Enterprise Institute stated in a 2025 analysis that most of Apple’s manufacturing supply chain remains in China, and the parts of it that have shifted to Southeast Asia and India tend to be the lower-value, more commoditized segments.¹⁷ The high-value, high-precision work, the kind that determines whether or not an iPhone 17 Pro ships on launch day, has not moved.

After India, Vietnam is the most developed of Apple’s alternative manufacturing bases. Apple has spent close to $16 billion through its Vietnamese supply chain since 2019, and the country now dominates AirPods and Apple Watch production. Vietnam’s advantage as an Apple manufacturer, however, is proximity to China—not independence from it.¹⁸ The vast majority of components feeding into Vietnamese assembly lines are still sourced from China.¹⁹ A serious disruption in China would, therefore, leave Vietnam exposed to the same underlying risks. Rising labor costs, worker shortages, and infrastructure constraints in Vietnam are now narrowing that country’s advantages.

The broader Southeast Asian picture is similar. Malaysia handles semiconductor packaging and testing, Thailand handles automotive-adjacent components, and Indonesia handles lower-end assembly. Each fills a niche and none can absorb iPhone-grade complexity at scale. All of them share the same structural dependency as India: roughly 90 percent of the components feeding Tata Electronics, India’s sole homegrown iPhone assembler, still originate in China.²⁰

Chinese suppliers have followed Apple’s diversification push into Vietnam and India, driven partly by commercial logic, partly by Beijing’s industrial policy, and partly because no alternative ecosystem yet exists capable of replacing them. Luxshare Precision, Goertek, and Lens Technology, three Chinese companies that are among the most important names in Apple’s supply chain, have all opened factories in Vietnam and now appear on Apple’s supplier list as Vietnamese manufacturers. Beijing has restricted Chinese Foxconn employees from transferring to iPhone factories there. Of Apple’s roughly 187 core suppliers, more than fifty are Chinese companies, and roughly 85 percent of all of its suppliers have factories inside China.²¹

The Roots of Apple’s Dependence

Apple’s own production strategy has deepened its China dependence. Chinese industry reporting describes Apple pressing suppliers to invest in automation to win or retain orders.²² In pressing for automation, Apple has effectively selected for the firms best positioned to do so. These are large Chinese manufacturers with state backing, deep capital reserves, and two decades of accumulated expertise: in other words, Foxconn and its most serious Chinese rival, Luxshare. Both now operate highly automated facilities running AI-driven quality control, predictive maintenance, and what Foxconn calls “lights-out” production lines (also known as “dark factories”), which are sections of a factory that run without human workers, monitored remotely through sensors feeding data to central management software.²³ Founded in 2004 by a former Foxconn worker, Luxshare won Apple’s AirPods contract by undercutting Taiwanese rivals on price while matching their precision; the firm has since been selected to manufacture Apple Vision Pro, arguably Apple’s most complex product. By 2024, Luxshare ranked first among mainland Chinese companies in smart precision manufacturing by revenue.²⁴

The infrastructure behind these capabilities cannot simply be relocated: it has to be rebuilt from scratch, and rebuilding it could take a generation. When Apple’s China dependence was primarily about labor costs, it was far easier to dislodge. The automation push has replaced that logic with something harder to escape: a lock-in built on capital, institutional knowledge, political pressure, and a supplier ecosystem that exists in China because Apple and its partners spent decades creating it.

The same logic applies when external pressure intervenes. When trade tensions rise, Apple consolidates around the suppliers it trusts to deliver at volume under difficult conditions. Most of those suppliers are in China. The same geopolitical and derisking pressures that are supposed to accelerate diversification end up reinforcing the status quo.

Take software as an example: Chinese regulations require AI services to obtain preapproval from the relevant governing body, the Cyberspace Administration of China (CAC), before they can be offered to users. Apple’s global AI product, built on OpenAI, has an uncertain legal presence in China and cannot operate there. Apple’s response was to find a domestic partner, and it reportedly considered Baidu, DeepSeek, Tencent, and ByteDance before settling on Alibaba. Alibaba’s chairman, Joseph Tsai, confirmed the deal publicly in February 2025, but Apple didn’t.²⁵

The cost of the absence is visible in the sales figures: iPhone shipments in China fell 17 percent for the full year of 2024—Apple’s largest annual decline in the country since 2016—with a 25 percent drop in the final quarter.²⁶ By year’s end, Apple had fallen to third place behind Vivo and Huawei.²⁷ On the Q1 2025 earnings call, Tim Cook told analysts that markets where Apple Intelligence had launched showed stronger iPhone 16 performance than markets where it had not. China was squarely in the latter column.²⁸

Both Apple and Alibaba submitted materials to CAC for approval. As of the time of writing, the rollout had slipped repeatedly, and Apple Intelligence remained unavailable on devices purchased in mainland China. The approval process itself is a form of leverage: once Apple has committed publicly to a partner, every quarter the feature does not ship is a quarter in which its premium devices fall further behind domestic competitors.

Washington’s reaction has exacerbated the problem for Apple. The Trump administration and members of Congress from both parties began publicly scrutinizing that partnership in the spring of 2025.²⁹ The House Select Committee on China and the House Permanent Select Committee on Intelligence asked Apple executives about data flows, commitments to Chinese regulators, and oversight mechanisms. According to the New York Times, the executives were unable to answer most of those questions. Representative Raja Krishnamoorthi (D-IL) described Alibaba as “a poster child for the Chinese Communist Party’s military-civil fusion strategy.”³⁰ Officials warned that the deal could hand Alibaba interaction data from hundreds of millions of iPhone users, the kind of real-world training material that improves AI models at scale.³¹ Officials in the administration reportedly discussed whether Alibaba should be placed on a restricted list barring U.S. companies from doing business with it.³²

Those concerns did not begin with Apple, and they do not end there. In November 2025, the Financial Times reported on a White House national security memo which alleged that Alibaba provided direct technological support to Chinese military operations targeting the United States. It also claimed the company supplied the People’s Liberation Army with customer data, including IP addresses, Wi-Fi information, and payment records.³³ Alibaba denied the allegations in full, calling the report a “malicious PR operation” designed to undermine Trump’s trade deal with Beijing. Three months later, in February 2026, the Pentagon added Alibaba to its 1260H Chinese Military Companies list: this is a formal designation that bars the Defense Department from contracting with listed firms, signaling to Pentagon suppliers and U.S. government agencies that the military considers the company a security risk.³⁴ The listing was withdrawn within an hour, without explanation, in what analysts read as a collision between China hawks inside the administration and a White House trying to protect the trade truce ahead of a planned Trump visit to Beijing. For Apple, the episode was a warning about the company it had chosen as its AI partner in China—and a demonstration of how quickly Washington’s posture toward any given Chinese firm can shift, with or without notice to the American companies that depend on them.

The Alibaba dilemma is the AI version of the same bind that defines Apple’s data governance in Guizhou and its manufacturing position in Zhengzhou: a dependency that is deeper than Apple’s public posture acknowledges, managed through ambiguity. These three exposures run in parallel, and each gives Beijing leverage over a core part of Apple’s business. Each draws increasing scrutiny in Washington, and none can be unwound quickly without significant cost or disruption.

Across his nearly fifteen-year-long tenure, Cook transformed Apple’s China relationship from a manufacturing arrangement into something closer to a strategic partnership with the CCP. He flew to Beijing repeatedly to meet senior officials, positioning Apple as a committed long-term partner in China’s technology economy. He deepened manufacturing concentration rather than hedging against it, binding Apple’s production capacity ever more tightly to Foxconn’s Zhengzhou complex and the dense supplier ecosystem surrounding it. When Beijing demanded governance concessions, he complied: iCloud data was handed over to Guizhou Big Data Group, VPN apps were stripped from the Chinese App Store, and the Taiwanese flag emoji was removed from Chinese devices. Each concession was defensible in isolation as the price of market access, but collectively, they amount to a strategic error that Cook’s successor will inherit.

The environment in which those concessions made sense no longer exists. Export controls on advanced chips have tightened, Chinese authorities are asserting greater control over data and AI services, and U.S. policymakers are increasingly treating corporate ties to China as national security issues rather than commercial ones. Ternus will face a genuine fork: commit to a managed decoupling that may take decades and which may be overtaken by events before it can be completed, or accept deeper integration with Chinese governance and absorb the consequences in Washington.

Apple’s Uncertain Paths

The most defensible path is managed decoupling—not the rolling deferrals and hedged announcements that defined Cook’s tenure—but a genuine strategic commitment to building a company that can absorb the reductions of China access without losing the ability to manufacture at the cutting edge.

For the last decade, administrations of both parties have treated the U.S.-China rivalry as a structural, generational contest, and the window for rebuilding outside China may not reopen after it closes. If Ternus moves early and accepts the near-term costs, he will be able to leave Apple in a position of genuine strategic autonomy. But this path is hard. It means absorbing years of margin pressure, manufacturing disruption, and political friction in both Washington and Beijing, with no guarantee the strategy pays off before a geopolitical shock, notably a Chinese invasion of Taiwan, forces the issue anyway.

In practice, this approach would amount to an industrial reconstruction process measured in years or decades, not quarters: rebuilding yield, engineering capacity, supplier clustering, and parts localization across several countries simultaneously. Some of that work has started: Apple now assembles a meaningful share of iPhones in India and has shifted AirPods and Apple Watch production to Vietnam (despite the China dependencies of these shifts discussed above). But the high-value work, such as advanced chip packaging, precision casing, and final assembly of flagship models, remains overwhelmingly concentrated in China, and no alternative ecosystem comes close to replicating it.

The deepest risk in this scenario is timing: the strategy works only if Apple can rebuild enough of the system before a geopolitical shock turns managed decoupling into forced decoupling. That shock will not necessarily come from China. The Iran War has illustrated how quickly apparently unrelated crises can converge on Apple’s supply chain. That war and the effective closure of the Strait of Hormuz since early March 2026 have put Taiwan’s energy supply under acute pressure. Taiwan imports roughly 97 percent of its energy, and a significant portion of its LNG flows through the Strait. TSMC alone produces nearly 90 percent of the world’s most advanced chips and itself consumes approximately 8 percent of Taiwan’s electricity.³⁵ Taiwan often has only eleven days of natural gas inventories in reserve, and an extended Strait of Hormuz closure or production stoppage by QatarEnergy, Taiwan’s largest single supplier of liquefied natural gas, could be especially damaging.³⁶ Apple’s chips are designed in California, made in Taiwan, and then assembled in China. Each node in that chain carries its own geopolitical exposure, and the exposures are not independent of each other.

Rare earths add a further layer that analysts rarely factor into the Apple-specific decoupling calculus. China accounts for approximately 60–70 percent of global rare earth raw material production, processes about 85 percent of the world’s output, and manufactures nearly 90 percent of the world’s rare earth magnets.³⁷ In October 2025, Beijing significantly expanded its export controls on rare earths and for the first time publicly asserted extraterritorial jurisdiction, requiring foreign exporters to obtain Chinese government approval to export items containing even trace amounts of Chinese-origin rare earths, including goods manufactured entirely outside China.³⁸ Those controls have since been partially suspended following bilateral negotiations, but the architecture remains in place and can be reactivated. Apple’s iPhones contain dysprosium in their vibration motors and a range of other rare earth elements in their displays and speakers. Moving the final assembly out of China does nothing to address that dependency.

Beijing has shown it can apply pressure at every point in Apple’s supply chain. Pulling Foxconn engineers from India required no law, no announcement, and no paper trail. Restricting rare earth exports required nothing more than a ministerial notice. Apple’s next CEO inherits a company exposed at both ends.

The second path is for Ternus to go the other direction entirely: this would mean deeper integration with Chinese governance, accepting that Apple’s China operations function under a distinct framework from the rest of its global business and betting that market access is worth what that costs in Washington.

Beijing filed a major antitrust complaint against Apple’s App Store in late 2025, targeting the 30 percent commission structure, the same rate charged by every major Chinese app store.³⁹ A formal ruling against Apple could cost billions in annual services revenue and force structural changes to Apple’s most profitable business line.⁴⁰ Under those conditions, a new CEO facing pressure on both Chinese revenue and Chinese manufacturing might calculate that formalizing the iCloud arrangements, locking in the Alibaba AI partnership, and accepting a permanently bifurcated governance model is the price of keeping the Chinese business intact.

Then, of course, Washington will react. In 2025 alone, Congress held fifty-six hearings on China policy, a 47 percent increase from the prior year, with technology and data security the dominant themes.⁴¹ The legal architecture for forcing structural separation from Chinese-controlled infrastructure already exists and has survived Supreme Court review. In 2024, Congress passed a law requiring ByteDance to divest or ban TikTok on national security grounds, and the Court upheld it unanimously in January 2025.⁴² The ban never actually took effect: TikTok continued operating in the United States, ByteDance retained ownership, and enforcement was repeatedly deferred despite the law standing and the Court having spoken. Compared to TikTok, Apple presents a different order of magnitude due to its centrality to U.S. technology infrastructure and its economic weight. The legal logic that impacted TikTok applies to Apple’s China arrangements just as readily. But the TikTok precedent suggests that the political will to act against a platform embedded in hundreds of millions of American lives may prove considerably weaker than the legislation implies.

Beijing, on the other hand, has never struggled to find the political will. No matter how much Apple localizes in China, the CCP has a long history of mistreating both foreign and Chinese businesses. Chinese businesses certainly have an advantage, which offers less protection than it might appear. When Alibaba founder Jack Ma made critical remarks about Chinese financial regulators in late 2020, Beijing halted the $37 billion IPO of Ant Group within weeks, then hit Alibaba itself with a record $2.8 billion antitrust fine.⁴³ Alibaba, a company that spent decades cultivating party relationships, saw its market capitalization fall by more than $600 billion.⁴⁴ Didi, China’s dominant ride-hailing platform, faced the wrath of Beijing just days after its June 2021 listing in New York, when regulators launched a data security investigation and removed it from app stores.⁴⁵

Foreign companies fare no better. The NBA made extensive concessions after the 2019 Hong Kong controversy—pulling broadcasts, silencing players, issuing apologies—and it still lost its Chinese streaming deal, spending the succeeding years trying to rebuild its Chinese audience. H&M published a statement declining to source cotton from Xinjiang, quietly retracted it under pressure, and still found itself wiped from Chinese e-commerce platforms and social media.⁴⁶ Years later, the Swedish multinational’s market position had not recovered.

Is There a Middle Way?

Even if Apple under Ternus successfully navigates the political pressures on both sides, satisfying Washington enough to avoid forced separation and Beijing enough to retain market access, it may find that the market it spent so much political capital protecting has already moved against it. In a July 2024 research note on Apple’s growth prospects in China, UBS analyst David Vogt warned that Apple was losing market share in China to local competitors. The trend is driven by stronger offerings from domestic brands and has resulted in the slowing of the global growth of iPhone sales.⁴⁷ Huawei, Xiaomi, and Vivo have spent years building AI ecosystems that are native to China, trained on Chinese data, and require no regulatory workaround to deploy. Apple may end up paying the governance costs without retaining the market.

There is a third path available to Ternus: maintain deep China ties, diversify just enough to satisfy regulators and investors, and manage both relationships through sustained political engagement at the highest levels of both governments. This is what Cook actually did, and for a long time it worked. But that strategy required something Cook had and his successor will not: time. Time to move slowly, to absorb each concession before the next one arrived, to let the relationship’s commercial logic outrun its political logic. That window has closed and the decisions that were deferred across two decades of extraordinary growth are now simultaneously due, in a geopolitical environment that is less forgiving, with less goodwill banked on either side.

Longtime Intel CEO Andy Grove argued that strategic inflection points are identifiable not by the data available at the time but by a shift in the underlying forces that determine what is possible. The data, he observed, is always about the past. By the time it confirms the inflection, the window for acting on it has often already closed. Apple is at that point now.

The question facing Ternus is what to do and where to go now that the firm’s China exposure has become so deeply rooted and structural to its basic operations, encompassing data governance, manufacturing, and AI simultaneously. Will Apple’s next chief executive treat this inherited position as a reason to act—rather than as a reason to wait for a clarity that will not come? Those CEOs who navigate inflection points successfully, Grove wrote, are the ones who move before the new reality is fully legible. The ones who wait for confirmation tend to find that confirmation arrives in the form of a crisis. Ternus will be judged, in the end, on a single question: did he see what the company had become inside the Chinese system, and was he willing to act while there was still time to choose?

This article originally appeared in American Affairs Volume X, Number 2 (Summer 2026): 90–102.

---

Notes